“I need to prove comprehension”
Experiences of literacy development of Swedish university students of Japanese
Jasmine Öjbro
Department of Asian, Middle Eastern and Turkish Studies Master's Thesis 30 ECTS
Japanese
Master's program in Japanese studies 120 ECTS Spring term 2016
Supervisor: Tore Nilsson
”I need to prove comprehension”
Experiences of literacy development of Swedish university students of Japanese
Jasmine ¨Ojbro Abstract
The aim of the thesis is to generate data for better understanding of the experiences of literacy development of students of Japanese as a foreign language. The informants consisted of students who were en- rolled in Japanese programmes at university level in Sweden. The re- search questions more specifically dealt with understanding students’
perceptions and experiences of reading in Japanese as well as per- ceptions on teaching methods of Japanese language at the university departments. The study also looked at expected course level profi- ciency and its relation to reading comprehension through a reading test. The methods of document review, interviews, questionnaire and were also used.
The results of the qualitative analysis indicate that students use their knowledge of grammar construction while reading to navigate texts and make sense of content. This method is more widely em- ployed by students who have received explicit instruction. Working actively with text seems to be beneficial for developing reading skills.
There is a shared perception among students that in a formal learning setting reading comprehension is the ability to translate idiomatically from Japanese. Many students experience difficulties with moving on from reading textbook material to native texts. Students also expe- rience issues with fluency and quickly recalling meaning and reading of words and kanji while reading.
A proposed improvement in programme structure for building flu- ency and bridging the gap between textbook and native material is to introduce the students to more reading material suitable to their proficiency level, gradually moving towards more advanced material.
Keywords: Japanese language, reading comprehension, literacy, for- eign language didactics, reading processes
ii
”Jag m˚aste bevisa f¨orst˚aelse”
Erfarenheter av litteracitetsutveckling hos svenska univer- sitetsstudenter av japanska
Jasmine ¨Ojbro Sammanfattning
Studiens syfte ¨ar att generera mer information och f¨orst˚aelse kring erfarenheter av litteracitetsuveckling hos studenter av japanska som fr¨ammande spr˚ak. Informanter bestod av studenter som var registr- erade p˚a spr˚akprogram f¨or japanska p˚a universitetsniv˚a i Sverige. Mer specifikt ber¨orde forskningsfr˚agorna att f˚a en f¨orst˚aelse f¨or studenters uppfattningar och erfarenheter av att l¨asa p˚a japanska samt upp- fattningar kring undervisningsmetoder f¨or japanska p˚a universitetens avdelningar. Studien granskade ¨aven f¨orv¨antad spr˚aklig f¨orm˚aga och hur det relaterar till l¨asf¨orst˚aelse. Metoder som anv¨andes inkluderar dokumentgranskning, intervjuer, enk¨at och l¨astest.
Resultaten fr˚an den kvalitativa anlysen indikerar att studenterna anv¨ander sina kunskaper om grammatiska konstruktioner medan de l¨aser f¨or att navigera text och f¨orst˚a sig p˚a inneh˚allet. Metoden anv¨ands mer extensivt bland studenter som har f˚att explicita instruk- tioner. Att arbeta aktivt med texter ¨ar f¨orm˚anligt f¨or utvecklandet av l¨askunskaper. Bland studenterna finns det en delad uppfattning om att l¨asf¨orst˚aelse i en formell l¨aromilj¨o inneb¨ar f¨orm˚agan att ¨overs¨atta idiomatiskt fr˚an japanska. Informanter upplever sv˚arigheter med att g˚a vidare fr˚an l¨asmaterial best˚aende av l¨arob¨ocker till material skrivet p˚a japanska riktat till modersm˚alstalare. Vidare upplever de problem med l¨asflyt och att komma ih˚ag betydelsen och l¨asningen av ord och kanji medan de l¨aser.
En f¨oreslagen f¨orb¨attring i undervisningen f¨or att utveckla l¨asflyt och ¨overbrygga hindret mellan l¨aroboksmaterial och japanska texter
¨
ar att introducera studenterna till l¨asmaterial anpassat efter deras niv˚a och att sedan gradvis n¨arma sig sv˚arare texter.
Nyckelord: Japanska, l¨asf¨orst˚aelse, litteracitet, spr˚akdidaktik, l¨aspro- cesser
iii
Conventions
Japanese text has been transliterated using the modified Hepburn sys- tem. A macron is used to indicate a long vowel. Transliterated words are written in italics and the accompanying gloss within single quo- tation marks (‘’). When necessary for context, the word is presented in its original Japanese orthographic form, either without or together with transliteration. There is only one gloss provided even if there are many possible translations. Words that are similar, or the same, in English and Japanese (e.g. kanji and kana) are treated as English words.
Transliterated Japanese names have been spelled in accordance with modified Hepburn, except in the case that a preferred translit- eration has been provided in context of the work. Japanese names are written in accordance with Western standards (given name + surname).
Titles of Japanese books are presented in italics in their romanized form with a footnote to a translation of the title the first time they are presented.
Changes to quotes are indicated using square brackets. Unless oth- erwise stated, no other changes have been made. Japanese quotes are presented in their original orthographic form with romanized translit- eration and an English translation. The translation is in normal script without quotation marks. Quotation in text is indicated by the use of double quotation marks and italics.
References are written in accordance with the Harvard system. Ti- tles of sources written in Japanese are presented transliterated with an English translation. Titles of sources written in other languages are presented in their original language together with an English trans- lation. In the instances where the authors have provided an English title for their work it has been used.
Italics are used for emphasis and referring to concepts. This paper follows British English spelling conventions.
iv
Acknowledgments
It is with deep gratitude that I’d like to acknowledge the following persons, for without their support writing this thesis would have been very daunting indeed.
Tore Nilsson, my most excellent supervisor, for invaluable feed- back, encouragement, and support throughout the entire process.
You’ve constantly provided me with motivation to strive for perfection and quality.
Christine Ericsdotter, my mentor, for setting me up with an of- fice space and providing indispensable advice on research and thesis writing. I hope I’ve returned all the books I borrowed.
Bj¨orn Forsberg, in-house mathematician, for guidance on all things statistical and making sure I got fed.
Maria, Olivia, Henrik, Adam, and Joel, for break-time with many cups of coffee and a side of smiles. Who would have guessed that taking breaks once in a while helps with the creative process?
I would also like to extend a heartfelt thank you to all the students who filled out the questionnaires and participated in interviews. Most importantly, my deepest gratitude to the Japanese departments at the three researched universities and their teaching staff who welcomed me with open arms.
v
vi
Contents
Introduction 1
1.1 Studying Japanese . . . . 1
1.1.1 Purpose of the study . . . . 2
1.1.2 Research method and analysis . . . . 3
1.1.3 Thesis outline . . . . 4
1.2 Central terms . . . . 4
1.3 Japanese language . . . . 6
1.3.1 Phonology . . . . 6
1.3.2 Vocabulary . . . . 7
1.3.3 Orthography . . . . 7
1.3.4 The Japanese-Language Proficiency Test . . . . 9
Literature Review 12 2.1 Literacy . . . . 12
2.1.1 Literacy and reading comprehension . . . . 12
2.1.2 Teaching reading and developing literacy . . . . 14
2.1.3 Lower-level processing . . . . 16
2.1.4 Higher-level processing . . . . 18
2.2 Language learning in higher education . . . . 19
2.2.1 Approaches to teaching . . . . 20
2.2.2 Approaches to learning . . . . 22
2.3 Reading Japanese . . . . 24
Methodological Considerations 26 3.1 Methodological basis . . . . 26
3.2 Research methods . . . . 26
3.2.1 Document review . . . . 27
3.2.2 Questionnaire . . . . 27
3.2.3 Reading comprehension . . . . 28
3.2.4 Interviews . . . . 29
3.2.5 Think-aloud . . . . 30
3.3 Overview of universities and respondents . . . . 32
3.4 Research procedure . . . . 33
3.4.1 Document review . . . . 33
3.4.2 Questionnaire . . . . 34
3.4.3 Pilot study and subsequent changes . . . . 34
3.4.4 Interview procedure . . . . 36
3.4.5 Pilot interview and subsequent changes . . . . 37
3.5 Analysis of empirical data . . . . 37
3.5.1 Thematisation of sentence completion . . . . 37
3.5.2 Valuing comprehension with probability . . . . 38
3.5.3 Systematic meaning condensation of interviews . . . . 39
3.5.4 Analysing the reading process during the think-aloud . . . . 40
3.6 Evaluation . . . . 41
3.6.1 Ethical aspects . . . . 41
3.6.2 Methodological consistency . . . . 41
Results and Analysis 43 4.1 Document review . . . . 43
4.1.1 Lund University . . . . 43
4.1.2 University of Gothenburg . . . . 44
4.1.3 Stockholm University . . . . 45
4.2 Questionnaire . . . . 46
4.2.1 Course level two . . . . 47
4.2.2 Course level four . . . . 50
4.3 Interviews . . . . 52
4.3.1 Abe . . . . 53
4.3.2 Bakhtin . . . . 54
4.3.3 Crystal . . . . 55
4.3.4 Dahl . . . . 57
4.3.5 Ellis . . . . 58
4.3.6 Fujitani . . . . 60
4.3.7 Student voices: Course level two . . . . 61
4.3.8 Student voices: Course level four . . . . 63
Discussion 64 5.1 Limitations . . . . 69
5.2 Implications . . . . 69
5.3 Future research . . . . 70
References 71
Appendix i: Questionnaire Results in Figures and Tables 77
List of Figures
1 Probability of Answering n Questions Correctly . . . . 39
2 Course Level Two - Frequency of Reading . . . . 77
3 Course Level Two - Reasons for Reading . . . . 78
4 Course Level Two - Reading Material . . . . 78
5 Course Level Two - Reasons for not Reading . . . . 79
6 Course Level Two - Reading Comprehension . . . . 82
7 Course Level Four - Frequency of Reading . . . . 83
8 Course Level Four - Reasons for Reading . . . . 83
9 Course Level Four - Reading Material . . . . 84
10 Course Level Four - Reasons for not Reading . . . . 84
11 Course Level Four - Reading Comprehension . . . . 87
List of Tables 1 Category of Vocabulary Words in the Japanese Language . . . . 8
2 JLPT: Summary of Linguistic Competence . . . . 10
3 Estimated Language Knowledge for the JLPT . . . . 11
4 Overview of Questionnaire Respondents . . . . 33
5 Japanese Knowledge and Abilities: Lund University . . . . 44
6 Japanese Knowledge and Abilities: University of Gothenburg . . . . 45
7 Japanese Knowledge and Abilities: Stockholm University . . . . 46
8 Course Level Two - Reading During Leisure . . . . 77
9 Course Level Two - Statements Regarding Study Habits . . . . 80
10 Course Level Two - Sentence Completion . . . . 81
11 Course Level Two - Completion Rate of Comprehension Test . . . . 82
12 Course Level Four - Reading During Leisure Time . . . . 83
13 Course Level Four - Statements Regarding Study Habits . . . . 85
14 Course Level Four - Sentence Completion . . . . 86
15 Course Level Four - Completion Rate of Comprehension Test . . . . 87
Introduction
1.1 Studying Japanese
It seems enough to enter the word Japanese into a search engine of choice to be met with a couple of hundred thousand hits, many of them having to do with studying the language. There is talk of a booming interest in most things Japanese, and the language itself, thanks to the spread of popular culture in the form of manga, anime and music. Japanese language schools are thriving and there’s a steady increase in students interested in learning Japanese (The Japan Times).
Another option closer to home for those who can’t study in Japan is the route of studying the language at university. Admission to Japanese courses and programmes at American and UK universities show continued interest and popularity. Sweden is no exception and yearly statistics for three of the universities offering Bachelor’s programmes in Japanese language studies show steady numbers of applications from prospective students.1 Students can either study Japanese as a single course or take part in the full programme which develops their language skills over the course of several consecutive terms.2
A typical language course often includes instruction on skills needed for language proficiency.
Depending on approach to language learning the syllabus can be formed to train the students in performing communicative activities (such as reading, listening, speaking, and writing) or teaching specific elements of the language system (such as vocabulary, grammar and language functions) (Robinson, 2009:295-296). Arguably no area is more important than another, but there is seemingly some interest in figuring out the specifics of second language reading in particular. Looking at current publications on the subject, the collected literature tends to focus on second language learners of English. Occasionally books on the subject may include chapters dealing with more non-commonly researched languages such as Chinese, Japanese or Arabic.3 Information about teaching second language reading in these lesser researched languages are often presented as summaries of case studies to further illustrate points raised about reading in English, or as further data on processes in general. Outside the context of reporting case studies they seldom get attention on their own.4 Concerning second language reading in general research deals mainly with either the specifics of the reading process and cognitive processing (see Grabe, 2009; Strømsø, 2007; Ehri, 1999) or how cognitive functions interact to help a reader extract meaning (see Alderson, 2000; Graesser et al., 2003). The first has to do with factual processes taking place as one reads in order for comprehension to occur. The second deals with literacy of readers and how reading can be instructed. Discourse on literacy focuses less on the actual processes used for understanding, and more on what comprehension requires in regards
1Statistics available online at Universitets- och h¨ogskoler˚adets antagningsstatistik (The council for university and college admissions statistics): http://statistik.uhr.se [In Swedish]
2See the referenced documents of syllabi from Lund University, University of Gothenburg and Stockholm University.
3For examples see Learning to Read Across Languages: Cross-Linguistic Relationships in First- and Second- Language Literacy Development edited by Koda and Zehler (2008), Reading in a Second Language. Moving from theory to practice by Grabe (2009), or Reading development and the teaching of reading: a psychological perspective edited by Oakhill and Beard (1999).
4For an illustrative example see Koda (2009) who describes several studies looking at reading in a second language, most of them ESL.
to understanding the language, the text, and the context (Alderson, 2000).
Reviewing studies concerning reading in Japanese in particular they may be concerned with the processes of learning specific aspects of processing Japanese. Dealing with learning how to read kanji or how words are processed, decoded and understood (see Kuwabara, 2009, 2010;
Koda, 2008; Konosu et al., 1998; Mori, 2014). These types of studies seldom make suggestions for teaching practices, and serve as illustrative examples of what constitutes processing Japanese in its written form. Reports and case studies describing effects of various pedagogical activities and efforts on developing areas of student reading proficiency are also common. These studies originating from a Japanese setting tend to deal with short term studies of smaller groups, often foreign exchange students who might only be enrolled in the language programme for a short period of time. They also differ in nature depending on area of interest. A study may deal with how to motivate students to study more, how to implement a successful reading course, or how reading material is perceived by students (see Kawakami, 2014; Hitosugi and Day, 2004;
Mikami and Harada, 2011, Tabata-Sandom, 2013). Commonly these studies are written from the perspective of teachers self-reporting on activities, often coupled with student opinions on the matter through questionnaires or short interviews. There is however a clear lack of long term perspectives on student development. I have been unable to find any comprehensive studies on the longitudinal literacy development of Japanese second language learners. It is not only of interest to know how novice students tend to study, but also how they study as they grow more mature in their language competence. Do their approaches change? To better understand the experiences of students while reading in Japanese, one also ought to look at the setting in which they learn the skills needed for processing, namely the institution providing a learning environment.
1.1.1 Purpose of the study
The aim of this thesis is to generate data to better understand long-term literacy development of students of Japanese as a foreign language in higher education in Sweden. The objective is therefore to map student literacy in relation to their own study efforts and perceptions of teaching within the frame of overall course learning objectives and curriculum design. This enables the identification of possible patterns of correlation between reading proficiency and areas such as course design or study methods. Furthermore, by mapping reading proficiency in relation to the expected course level proficiency the student is supposed to have acquired, as opposed to actual individual language performance, greater knowledge of how course level corresponds to a baseline for language proficiency is obtained. The research questions are:
1. What is the relationship between student literacy and expected course level proficiency?
2. What are students’ approaches to reading in Japanese?
3. What are students’ perceptions of overall teaching methods of Japanese language?
4. What are students’ perceptions of the experience and process of reading in Japanese?
1.1.2 Research method and analysis
This thesis builds upon data collected from Japanese language programmes offered at three universities in Sweden: Lund University, University of Gothenburg and Stockholm University.
These three in particular were chosen from a multitude of academic institutions due to their intra-comparability. All programmes focus on the Japanese language and offer language courses culminating in a Bachelor’s degree in Japanese language studies. After completing their degrees students have access to postgraduate studies at the same departments. At the time when the research was carried out all three departments had scheduled instruction for course levels two and four. By surveying both course levels the objective was to collect data that could be used as the basis for predicting and discussing student development.
In order to answer the research questions and investigate the identified research gap from different angles the study employed a range of methods. The approach is phenomenological and uses interviews and surveys for data collection which yielded both quantitative and rich qualitative data. The methodological considerations for each part of the study as well as the research procedure and analysis are described in full in section 3.
The study itself was conducted in three steps. The first step consisted of a document review summarising and analysing syllabi of the three departments to determine teaching approaches and required knowledge for each course level part of the Japanese programmes. Analysis for the first part was admittedly limited in scope as the resulting interpretation was not further checked by representatives of the departments. Syllabi documents were taken at face-value and summarised at the researcher’s discretion. The review thereby reflects general assumptions one can make regarding the programme and may in fact not represent the intentions of document writers. The aim of the review was to gain an understanding of the curriculum and its methods in relation to how students may interpret it in order to put later data from respondents into perspective. This information was used in developing the questionnaires and to better relate research findings and interpret student perceptions.
The second step surveyed students from the three departments. Groups from course levels two and four answered questionnaires about their reading and study habits as well as on their perceptions of reading instruction. They were also asked to complete three reading comprehen- sion tasks. The intention was to see if there were discernible patterns between schools due to syllabi construction and instruction and if there were commonalities among proficient readers.
The collected data was both qualitative and quantitative in nature, giving some indication as to how students study, what and why they read as well as reading strategies. Besides frequencies of different choices analysed in SPSS, the results of the questionnaire also entailed thematising data from sentence completion tasks and valuing the results of the comprehension tasks. The results of the questionnaire gave insight to some of the research questions.
The third step of the study consisted of six semi-structured qualitative interviews with one respondent from each course level at the three departments. During the interviews the subjects were asked to complete a reading task and to speak about a range of topics including their perception of current teaching methods and difficulties with reading Japanese text. Some questions were used to clarify data from the first two parts of the study. The summarised
interviews are presented in section 4.3 together with a combined narrative that represents the two course levels (levels two and four) as a whole. The results of the interviews and reading tasks gave further insight into some of the research questions.
Pilot studies for the survey and interviews were conducted with a small selection of students enrolled in the final term of the Bachelor’s programme. Their involvement helped shape the final survey and interview process but the results were not analysed or part of the final presented data.
1.1.3 Thesis outline
The first section consists of two halves. First, the information already presented which provides an overview of the research gap, purpose of the study and its research question as well as chosen method for gathering and analysing data. This is followed by an introduction of central terms and how they are used throughout the paper as well as the theoretical assumptions that they are composed of. A general guide to the Japanese language is also provided alongside related concepts which are of importance later in the literature review and the section on methodological considerations.
The following section deals with literature review and aims to guide the reader through the relevant previous research. Starting with literacy and reading comprehension the section moves on to explain the reading process in general and provides some examples relating to Japanese.
The theme of language learning in higher education is then presented. The theme takes a step back from Japanese in particular and presents different theoretical perspectives of how syllabus design and approaches to teaching affect students and learning outcomes. Research and theories presented within the subjects of literacy and language learning in higher education make a reappearance in the discussion of research results. The final section is dedicated to the theme of reading Japanese; collecting studies and research on processing Japanese language and practices for teaching reading.
Following the literature review is a presentation of the methodological considerations ob- served when designing the study. It goes into detail about the research procedure and methods used for collecting and analysing data. There is also some discussion of ethical aspects and methodological consistency. The section titled Results and Analysis presents the research find- ings. The three parts of the study are presented under their respective categories with related tables and figures collected in Appendix i. The last section consists of a discussion of the results. The results are discussed holistically in relation to theory and previous empirical re- search, attempting to see the wider implications of the research findings. This part also deals with limitations of the study and makes suggestions for future research.
1.2 Central terms
The term L1 refers to the first language acquired by a speaker. The term for languages acquired after the native language is L2, referring to both second and foreign language, a distinction based on environmental circumstances. Second language (SL) most often refers to a language learned as a speaker is living in an environment where the target language is widely used. Foreign
language (FL) on the other hand is when a language learner is acquiring the language in an environment outside of the target language, often in an instructional setting. Second language is used as an umbrella term for both second and foreign language. Target language (TL) refers to the language a person is studying in the hopes of attaining.
The two terms acquisition and learning are often used to distinguish between two related concepts of gaining language ability. Acquisition is a highly automated process where language aspects are absorbed incidentally. This type of implicit knowledge builds upon competence associated with internalised procedures and knowing how. Learning on the other hand is often associated with a conscious process of learning about the language itself (Paradis, 2004:7-10;
Ligtbown and Spada, 1999: 177). While some of the reviewed literature makes a distinction between the two terms, I use the terms interchangeably later on.
Process, production, performance, proficiency, and competence
When discussing language it is vital to distinguish language process from language production (Alderson, 2000:3-6). Process deals with the interaction between a speaker and the language, the mental processing that takes place when registering and interpreting language. Production on the other hand is the concrete language output, i.e. what a speaker says. To complicate things further, process may be used to refer to a chain of events of either mental processing or language production. Used when speaking of literacy development the process of reading could then refer to either (1) the mental processing of interpreting language and the written word, or (2) the act of reading and what transpires when one develops literacy e.g. gradually reading more difficult texts. There are several key concepts related to processing and production. The terminology in question labels aspects within the spectrum of language proficiency, but there is no consensus on definitions. Regarding definitions the thesis uses the following:
Performance is the same as output and is directly related to language production and the actual usage of a language (Lightbown and Spada, 1999:178). Performance does not take input or comprehension into consideration, although it is assumed that performance requires some form of comprehension of what is being produced.
Proficiency is complex to define. Scholarly discourse is divided but typically includes accu- rate language comprehension and production alongside fluency in a number of communicative situations (Del Vecchio and Guerrero, 1995). Language proficiency tests for example focus on determining ‘how much of the L2 a learner has mastered ’(Crystal, 2010:397). Proficiency can therefore be defined as the extent of L2 knowledge.
Competence is the mental knowledge a speaker possesses of a language (Paradis, 2004:7-10) and an underlying feature of language processing. Consisting of explicit and implicit knowledge of the language it is the result of both learning and acquisition. They work interchangeably to provide the speaker with various information; where the implicit knowledge is insufficient explicit knowledge can fill in the gaps, and vice versa (Paradis, 2004: 10-12). Competence can be further divided into smaller components for distinguishing between related abilities. Linguistic competence for example includes knowledge of linguistic structures, syntax and pronunciation (Whong, 2011:70).
Factors in language learning
Research on language acquisition often mentions factors that may have an impact on ultimate language attainment. Factors such as motivation, aptitude, attitudes, and exposure to the language play an important role. The following paragraphs deal with factors affecting language acquisition.
Aptitude is an innate skill for processing and acquiring language structure. As presented by Lightbown and Spada (1999:53-54) it entails rapid learning of language features thanks to abilities in identifying and memorising language features. Aptitude entails several types of abilities, differing between persons. It is part of the individual make-up of inherent skills upon which further language learning and acquisition build. Aptitude can be measured, but it is a rather lengthy procedure which serves as poor prediction of ultimate language attainment (Lightbown and Spada, 1999:54).
Motivation plays a key role, either as the grounds for engaging in language studies or pro- viding reasons to propel them forward. However, research has yet to identify the exact role of motivation within the broader frame of language learning; it is unclear whether motivation leads to successful learning or if successful learning produces motivation (Lightbown and Spada, 1999:56). However, studies mapping the role of motivation on language attainment have identi- fied several types of motivation that stimulate the learner in developing her L2 (Biggs, 2003:61;
Kumaravadivelu, 2006:40-41; Entwistle, 2009:20). A few studies presented in section 2.3 mention setting up reading programmes for Japanese language students that have an effect on student motivation to continue reading.
The factor of motivation in itself is rather complex and is affected by other factors such as environment, affective reasons and individual circumstances. However, two highly cited types of motivation that are important to mention as part of the broader concept are intrinsic and extrinsic motivation. The first type derives from an interest in what is being learned, acting as incentive to continue developing, while extrinsic motivation depends on external factors such as obligatory school work (Entwistle, 2009:20).
1.3 Japanese language
This section serves as a general introduction to the specifics of the Japanese language system.
The information presented here serves as the basis for understanding research presented in the literature review. The connection between the three areas of phonology, vocabulary and orthography are later discussed within the subjects of reading comprehension and literacy. The Japanese Language Proficiency test is explained since it serves as basis for the questionnaire, more specifically the choosing of texts for the reading comprehension.
1.3.1 Phonology
Japanese phonemes consists of 5 vowels (a, i, u, e, o) and 16 consonant-sounds (Taylor and Taylor, 1995: 283), there are even more if you count different realisations and dialects (Tsu- jimura, 2007). The five vowels are phonemes and mora in their own right, and combined with
consonants they form the bulk of Japanese sound inventory. Consonants are always within a CV-pairing5, with the exception of a nasal /n/, and there are no consonant clusters.
The Japanese language make us of rythmic units known as mora, which is a unit somewhere between a segment and a syllable. An example provided by Tsujimura (2007:58) uses the word London as an example. The English speaker would divide it into two syllables, Lon.don, while the Japanese speaker hears four, Lo.n.do.n. The notion of mora is important for grasping the structure of the language in terms of phonemic inventory and written representation.
1.3.2 Vocabulary
As presented by Taylor and Taylor (1995:284) the Japanese vocabulary consists of words from four different origins: Japanese native, Sino-Japanese words (historical borrowings from Chi- nese), foreign loan words (for example: Portuguese, Dutch and English origin), and hybrid words (stems of Sino-Japanese or foreign loans with Japanese endings). A distinction between these types is necessary since they are usually written with different kind of scripts. Sino-Japanese words for example are almost exclusively written in kanji while hybrid words commonly use a combination of kanji and kana. An overview is presented in table 1.
Sino-Japanese words trace their origins back to the importation of kanji script from China to Japan (Lindberg-Wada et al., 2006). While Japanese native morphemes are inclined to be multi- syllabic in their construction, Sino-Japanese morphemes on the other hand are monosyllabic. For example the Japanese native one-morpheme word hana ‘flower’, has a two-syllabic structure, while the Sino-Japanese one-morpheme word is monosyllabic, ka ‘flower’(Taylor and Taylor, 1995). The relationship and history between Sino-Japanese and Japanese native words can also be seen in the word usage within certain categories. Native words tend to be used in both oral and written communication for everyday activities such as verbs and nouns while Sino- Japanese words are frequently used for abstract concepts, institutions or scholarly disciplines (Taylor and Taylor, 1995: 286). Furthermore, as Sino-Japanese words are derived from their Chinese counterpart and imported together with kanji script they are most often written with the corresponding kanji.
Foreign loan words are newer introductions to the Japanese vocabulary than those of Chinese origin and they have oftentimes changed form or meaning drastically (Taylor and Taylor, 1995).
Loan words are usually represented in writing using katakana, as they have been adapted to fit the Japanese phonology. An example being the Portuguese word carta becoming karuta ‘playing card’. Several loan words may appear similar but have changed meaning, such as the English example stove becoming sut¯obu with the new meaning ‘electric heater’.
1.3.3 Orthography
Japanese employs kanji, a complex orthographic system of units representing semantic meaning and morphemes, and the two syllabic scripts hiragana and katakana, usually referred to as kana.
Although making use of the Roman alphabet on occasion as well as Arabic numerals, these two
5C represents consonant and V stands for vowel.
Table 1: Category of Vocabulary Words in the Japanese Language
Category Example Japanese Type
Native kawa (‘river’) 川 Kanji
Sino-Japanese gakka (‘department’) 学科 Kanji Foreign terebi (‘television’) テレビ Katakana Hybrid ai-suru (‘to love’) 愛する Kanji-hiragana
Example words, translations and scripts for the four category types of Japanese vocabulary words.
Examples from Taylor and Taylor (1995:284).
systems will not be covered extensively in this section as readers are supposedly somewhat familiar with them.
The kana systems are connected to the previously mentioned Japanese moraeic structure of language and correspond to moraeic utterances of the V/CV-structure. Hiragana is a rounder, seemingly softer, script whose form derived from the kanji-system and is used extensively for native words and as grammatical markers in kanji-kana combinations. For example, a verb stem may be represented by a kanji while the suffix of the verb ending is written in hiragana, known as okurigana6. Katakana on the other hand has a more angular visual representation extensively used for foreign words such as personal names, place names, loan words and onomatopoeia.
Hiragana and katakana both consist of identical V/CV-structures, but with distinct orthographic representation, where each script consists of 46 syllabograms (Lindberg-Wada et al., 2006).
In kana, a couple of syllabograms share an identical visual representation (within the hira- gana/katakana scripts) with the addition of diacritic mark (two dots or a circular shape) that indicates adding sound to a consonant. Examples includeは ha turning into the voiced plosive ば ba with the addition of two dots, and the voiceless plosive ぱ pa forming with the added circle. Further writing conventions include combining the CV-structure where the mora ends in an i, with a visually smaller y-mora (ya, yu, ye) to create a diphthong-like pronunciation.
Furthermore, both kana make use of a smaller version of the つ/ツ tsu character before the following mora for indicating that the following consonant in the CV-structure ought to have a more marked pronunciation (Taylor and Taylor, 1995:283; Lindberg-Wada et al., 2006).
Kanji were initially brought to Japan in the 4th century AD, and firstly used in accordance with Chinese writing conventions (Taylor and Taylor, 1995:295; Lindberg-Wada et al., 2006).
The writing did not perfectly match Japanese words and in the beginning they were used as phonemic representation oftentimes regardless of their semantic meaning. Over time kanji were adapted in order to fit Japanese language conventions better, with clearer connections between both phonemic and semantic representation. Today, kanji have been changed, simplified, and new ones have been invented to create a functioning orthography (Lindberg-Wada et al., 2006:).
Kanji are often simply referred to as a logographic script, meaning that a sign represents a word.
This does not hold entirely true as only a handful of signs are pictograms, representing concrete objects. The bulk of the kanji system is a potpourri of several types of signs combining semantic
6Okurigana are suffixes of kana following a stem written in kanji. Okurigana is only used for kun’yomi words of Japanese origin, not for on’yomi words of Chinese origin. Example: 見る ‘see’where 見 is the stem and the kanaる is the okurigana.
meaning and phonological representation in several ways. Apart from pictograms or ideograms, kanji have developed through a variety of ways: linking semantic meaning of pictograms to create a signifier for a related term (月, ‘moon’, and 日, ‘sun’, combine to form 明, ‘bright’), compounds consisting of varying constellations of radical (indication of meaning) and phonetic (pronunciation) elements (休, ‘rest’consisting of 人, ‘man’, and 木, ‘tree’), and attaching native morphemes to a character of similar meaning or sound (Lindberg-Wada et al, 2006:14-15). The number of kanji used in Japanese are in the thousands, and not all of them are part of compulsory education. The so-called J¯oy¯o kanji is a list of the 2,136 kanji taught in Japanese schools up to secondary school-level, serving as a baseline for literacy (MEXT, J¯oy¯okanji-hy¯o).
As mentioned above, kanji is not a strictly logographic script as signs represent mora rather than solely carrying semantic meaning. Furthermore, they can have several meanings depend- ing on usage e.g. when combined with other kanji to form compound words, as well as several pronunciations. Some morphemes associated with kanji trace their origins back to the impor- tation of the script itself, where they retain a Japanese approximate pronunciation of Chinese words. Readings of Sino-Japanese kanji are called on’yomi ‘sound based reading’. Another possible reading of kanji is the kun’yomi ‘meaning based reading’, representing the morphemes of Japanese native words (Lindberg-Wada et al., 2006). Single kanji, often written in combi- nation with okuriana for adjectives or verbs, tend to be read with kun’yomi while two kanji or longer compound words tend to be read using the Sino-Japanese on’yomi (Taylor and Taylor, 1995:301). As there oftentimes are several associated on’yomi readings of a kanji, compound words may be realised in one of several ways depending on the combination of on’yomi and kun’yomi. For example, the word 生物 can be read with two on’yomi to form the word seibutsu
‘living thing’, or two kun’yomi to form namamono ‘raw food’.
1.3.4 The Japanese-Language Proficiency Test
The Japanese-Language Proficiency Test (日本語能力試験, Nihongo N¯oryoku Shiken) commonly abbreviated as JLPT, is a standardised test for non-native speakers of Japanese. Evaluating language proficiency within reading and listening abilities as well as general language knowledge, the test is divided into five levels. Each of the levels corresponds to an expected level of language command in terms of vocabulary, grammar and kanji. The basic level is called N5 and the most advanced N1, and the five tiers measure proficiency in Japanese through listening and reading tests. The JLPT has great importance when applying for work or a visa, as those who hold a certificate of clearing N2 or N1 may be privy to speedier visa expedition, entitlement to a raise, or employment. Therefore, JLPT can be seen as the yardstick for what foreign speakers of Japanese are expected to master, and it is a tool for comparison of linguistic competence.
In the case of reading, the JLPT can be seen as a standardised measure for level of reading proficiency.
Concerning reading proficiency of the JLPT, each tier of the five levels has a set of linguistic competence abilities necessary to pass the test for that level. The abilites are in turn expressed as the ability to understand and properly interpret certain real-life material such as newspaper articles, textbooks and prose. Summarised in table 2, the basic and elementary levels measure
Table 2: JLPT: Summary of Linguistic Competence Level Summary of Linguistic Competence
JLPT One is able to read and understand...
N5 - Typical expressions and sentences - Hiragana, katakana, and basic kanji N4 - Passages on familiar topics
- Basic vocabulary and kanji
N3 - Specific contents concerning everyday topics - Summary information
- Newspaper headlines
- Slightly difficult writings in everyday situations - When aided by alternative phrases
N2 - A variety of topics
- Articles and commentaries in newspapers/magazines - Comprehend main content
- Intent of the writers and follow narratives N1 - Complex topics
- Newspaper editorials and critiques - Structure and content of texts
- Intent of the writers and follow narratives comprehensively
A summary of the linguistic competence necessary to be considered proficient at the five levels of the Japanese Language Proficiency Test (JLPT, N1-N5: Nintei no meyasu).
Japanese proficiency corresponding to a basic understanding of texts on the word and sentence levels. They involve understanding texts and conversations dealing with familiar everyday topics.
The intermediate level entails an ability to understand Japanese used in everyday situations with slight more complexity than the previous two tiers. For the upper-intermediate and advanced levels a speaker is expected to comprehend a great deal of written and spoken material in a variety of topics. They require heightened literacy in terms of interpreting underlying meaning and essential points. Advanced levels require a deeper understanding of texts in a broad social context. Further guidance as to sub-skills necessary for passing the tests is available through official preparatory books. They present the kanji, vocabulary and grammar knowledge covered in each level that may be on the test. Table 3 shows an overview of the estimated number of kanji and vocabulary required for each level of the JLPT as well as recent data on the hours of study students usually spend in order to reach proficiency within that tier.
The current division of the five-tier JLPT was first introduced in 2010. Previously the test consisted of only four levels. The current N3 was added to the structure in order to bridge a considerable gap between the (then) levels of 3 and 2. In other words as the N3 was added in 2010, with the linguistic competence needed for the other four tiers largely unchanged, the two lower tiers (3 and 4) were renamed into N4 and N5 respectively (JLPT, Ky¯ushiken to no hikaku.).
Table 3: Estimated Language Knowledge for the JLPT
Level JLPT-level Kanji Vocabulary Hours of study
Basic N5 100 800 325-600 hours
Elementary N4 300 1,500 575-1000 hours
Intermediate N3 650 3,750 950-1700 hours
Upper-intermediate N2 1,000 6,000 1600-2800 hours
Advanced N1 2,000 10,000 3000-4800 hours
An approximate estimation of the amount of kanji and vocabulary knowledge needed for each level of the Japanese Language Proficiency Test. The estimated hours of study is based on study hour comparison data of students from a dissimilar orthographic background (The Japanese Language Education Center, JLPT Study Hour Comparison Data 2010-2015).
Literature Review
2.1 Literacy
This section looks at the term literacy, further defining it in relation to reading proficiency and presenting theories on the development of literacy as well as the mental processing of reading.
Starting with theories related to the nature of reading in a first language, it also deals with the general framework of second language reading, comprehension and the measurement of reading proficiency.
2.1.1 Literacy and reading comprehension
Literacy is a broad term that may encompass several aspects of written and oral communica- tion. Scholarly discourse highlights understanding; literacy is seen as a move beyond merely decoding individual letters and entails using cognitive reasoning to put the meaning of words into context or infer implicit information in order to understand the text as a whole (Alderson, 2000:9; Graesser et al., 1997:178). In other words literacy is the summation of cognitive and linguistics skills yielding a reader capable of comprehending the written word on several cog- nitive levels. Looking further at the two word comprehension and literacy, comprehension is a requisite for literacy. Comprehension equals understanding to some extent through different mental processes. Literacy takes a step further and puts comprehension into a broader social perspective.
Regarding the term comprehension Grabe (2009:22) explains that reading comprehension is a process involving higher and lower level processing. These are both linguistic and cognitive in nature and combined help a reader understand. Lower-level processes need to be somewhat automatic, the more smoothly they operate without conscious thought (in connection with other factors) the more likely comprehension will occur. These processes, or component abilities are basically language-processing skills and include word recognition, vocabulary knowledge, using grammatical information and building meaning from the collected information of word meaning and grammatical information (Grabe, 2009:36; Strømsø, 2007:55-57). Each component ability is in turn made out of cognitive processes. For example decoding is a vital part of word recognition where readers use their understanding of a language’s phonology and orthography to identify individual letters, translate them to word-sounds and comprehend individual words (Ehri, 1999).
It is when looking at higher-level processes as a concept that scholarly discourse diverges and alternate terms between reading comprehension and literacy. To be more precise, scholars agree that lower-level processes are the foundation for literacy and comprehension. Readers gain read- ing proficiency through better implementation of the lower-level processes (see Alderson, 2000 and Graesser et al., 2003) as it helps them grasp the very fabric of a text. The divergent terms comprehension and literacy speak of the actual understanding of the textual meaning when higher-level processes are involved. There is still ongoing debate as to what exact cognitive processes are involved in this step and what model can be used to describe the entire compre- hension process. The basic point of agreement seems to be that higher-order comprehension involves processes that bind linguistic information from lower-level processes together with the
readers understanding of context and discourse. This can involve anything from having back- ground knowledge of situations described in the text which aids understanding its main points to inferencing information from textual cues. Especially literacy focuses on how successfully a reader can navigate and understand a text using these higher-level comprehension processes and frequently speak of levels of understanding. Stuart et al. (2009:56) summarises literacy as such:
It follows that in every act of reading, once word meanings are activated in the language system, semantic and syntactic processes (combined with the reader’s rel- evant real-world background knowledge) allow the reader to start to determine the literal meaning of the sentences and to build a mental representation of the situation described in the text.
Alderson, from an example by Bransford, explains that even simple sentences require language processing to make inferences in order to understand: ”The floor was dirty because Sally used the mop” gives the inferenced meaning ”the mop was dirty” (Alderson, 2000:8). A literal in- terpretation misses this statement. It has been suggested that during higher-level processing readers construct a mental representation of a text on multiple levels (Graesser et al., 2003:10).
Certainly, several components contribute to overall comprehension. As previously mentioned some of these are background knowledge and the ability to make inferences. Payne (2009:119- 120) takes a further step and claims that depending on reading context readers comprehend to different extents. This reasoning looks at literacy as something domain-specific, where readers can become more proficient at comprehending certain types of material. For example frequently reading newspapers equips readers with component abilities specifically needed for that genre.
More familiarity with related vocabulary or an understanding of sentence structure and narra- tion of articles might lead to more fluency and being able to grasp content more quickly. This is higher-level processing at work. Thusly, literacy is highly context dependent. A difference in comprehension between text types often stems from a lack of conceptual understanding of genre or previous knowledge against which the text is interpreted (Strømsø, 2007: 68). It is not entirely difficult to envisage that a skillful reader may be more proficient at comprehending certain types of text through extensive exposure and practice.
A model for building mental representations during reading, as proposed by Graesser, consists of five levels of understanding: (1) the surface code which is the grammar and wording of sentences, (2) textbase that is the explicit meaning of the text as well as small inferences necessary for coherence, (3) situation model which is the content of what the text is talking about, (4) communication level of the ideas communicated through the text, and lastly, (5) the structural components of the specific text genre (Graesser et al., 1997:167; Graesser et al., 2003:10-11).
The first two components roughly translate to lower-level processing while the other three are of higher-level cognitive processing. Graesser further states that in order to become a proficient reader, reaching comprehension and having deep understanding of a text, one should have some mastery of each component.
As I see it, the discussion on processing relates to what components are necessary for com- prehension. Literacy’s levels of understanding relates to how well these components interact in order for the reader to make successful inferences and interpretations on which comprehen-
sion is based. Using the previous example of the dirty mop, higher-level processing results in inferencing this meaning while literacy is the continued success of inferencing such statements throughout the entire text. Not having deep levels of understanding and lacking in literacy does not mean that high-level processing is absent. It merely indicates that there are gaps in the cognitive data used for these processes in comprehending.
There are three prevalent views on how higher-level and lower-level processing interact while reading. One of the earliest frameworks is the idea of a bottom-up approach meaning that a reader reads a text by serial decoding starting with the smallest elements: seeing the ortho- graphic elements, decoding them into speech-sounds, understanding the words they represent and from this infer meaning (Alderson, 2000). The approach looks at the processing of reading as individual skills activated in sequential order where the whole is understood through the summation of the parts (Stanovich and Stanovich, 1999:24). This model has influenced edu- cational efforts to focus reading instruction on individual components. The skills are taught through direct explicit instruction and make frequent use of rote learning (Rassool, 2009:9-10).
The reading process and literacy development is seen as a serial schema where one tier of skills builds upon another. In other words, as reading is the summation of many parts one must build upon all of the parts individually in order to progress as a reader. This framework of explanation has been criticised for not fully addressing the influence of higher-level processes of mental representations and extracting the meaning of texts (Rassool, 2009:11).
The top-down approach on the other hand is an opposite model to the bottom-up approach.
It advocates that the process and production entail making sense of texts mainly by relying on prior knowledge (Rassool, 2009; Alderson, 2000). The decoding process of graphemes7to sounds to meaning is only employed when the reader comes across unfamiliar elements not predicted by their internal schema. Texts are understood from whole to parts, and the approach heavily emphasises the connection between reading process and context needed for a more complex understanding of literacy in a social perspective. However, critics have noted that there is a lack of studies supporting concrete evidence for this line of process-production (Rassool, 2009:9-11).
A third emergent frameworks look at literacy as a parallel process where decoding and cogni- tive processing take place simultaneously, feeding into each other and making the reader adjust her understanding of the text as reading progresses (Alderson, 2000:15-18; Farrall, 2012:17-26;
Grabe, 2009:55). It can be seen as a merger between the top-down and bottom-up approaches, acknowledging that a beginner reader becomes more skilled as their language competence grows while giving recognition to the importance of higher-level processing abilities which goes on at the same time.
2.1.2 Teaching reading and developing literacy
Much of the reading and literacy discourse focuses on emerging readers in a first language and the perspective of how children first acquire abilities pertient to reading such as mastering the decoding process or learning new vocabulary. The point of interest is pedagogical (how do we teach this? ) rather than cognitive (what happens while reading? ) and deals with the
7Graphemes are the minimal units of a writing system, in this case individual letters representing sound
