Why English learning fails in Japan: Reason #3

Why English learning fails in Japan: Reason #3

Note: This is the third of a series that will run for at least ten parts. However, readers' comments are welcomed and, where relevant, will be added as reasons in the main series, with full credit given.

Reason #3: Japanese is not written with an alphabet


Japanese is not written with an alphabet.

The writing system used to represent Japanese in continuous text is notorious for its complexity. Chinese characters (called kanji in Japanese) have been adopted and adapted to write Japanese. But one problem with Chinese characters is that they don't fit nicely onto Japanese. Chinese is a group of languages characterized, in part, as being 'mono-syllabic'. That is, one content morpheme of one spoken syllable in length is usually written in text with one character. However, Japanese is, unlike Chinese, highly inflected and multi-syllabic. Using Chinese characters to represent Japanese leads to syllabic and lexical opacity--one character might stand for one, two, three or more spoken syllables.

To make matters worse in terms of the complexity of written Japanese, Chinese characters are typically used to stand for different words or morphemes that may (or may not) cover some of the same semantic areas. For example, the same character might be used to represent two words or morphemes that are synonymous, even though their pronunciations are totally different. This explanation only hints at some of the complexities in adapting such a mismatched writing system to Japanese.

As written Japanese evolved, the Chinese writing system was modified and enhanced to make it more useful for full-blown Japanese texts. In order to help solve the opacity issue, two syllabaries have been put to use to complement the use of Chinese characters. These are called hiragana and katakana.

As these sets of written syllables are used today in written Japanese, hiragana (a cursive set of syllables) is used to write much of the multi-syllabic inflections and particles of the language. It is also often preferred over the use of the kanji for many high-frequency, everyday words and phrases. Clearly, the idea of using a word-level graph (such as a kanji) to write a language comes to Japan from China. But the use of syllabic graphs might come from India through the diffusion of literacy practices associated with Buddhism.

The other syllabary, katakana, captures in written form the same simplified, idealized syllables abstracted from spoken Japanese and captured in hiragana, but it is written with elements derived from kanji and displays the same sort of angularity. Katakana is typically used to write foreign loan words, especially non-Chinese ones. It is also used to represent the elements of Japanese vocabulary that imitate aural, other physical, and mental phenomena (what is called roughly 'onomatopoeia' in English, though it is much more extensive and developed in Japanese). It can also be employed to show emphasis, analogous to the way all caps or italics might be used in English.

Full-blown written Japanese will typically show mostly kanji and hiragana, with a smattering of katakana (unless the text was about foreign loan words, or was a menu at a western restaurant). A text of Japanese might also have quite a number of items in the roman alphabet (called romaji in Japanese) and Arabic numerals (though Chinese numerals are also often used).
But roman letters are rarely used to romanize Japanese, except in the case of synonyms like 'UN' or 'NATO' (which do have their own Japanese pronunciations in the spoken language). Romanization does happen in advertising and on shop signs. You are most likely to see words written in the roman alphabet in academic discussions where an English or other foreign word or phrase is placed in the text and then explained in Japanese.

Since most Japanese--the sort of language that would be spoken informally but also formally--is rarely put into alphabetic form (romanization). Most native speakers of Japanese are literate in their own language, which is largely not alphabetically written. They lack much familiarity or fluency with the writing and spelling conventions of English.

For example, most university students will struggle with a dictionary laid out in traditional alphabetical order (A to Z) because Japanese language dictionaries are typically put into a syllabic order. And one hindrance to personal computer use in Japan has been learner reluctance to interface with a computer using an alpha-numeric keyboard that looks totally alien in terms of the salient features of Japanese literacy. In other words, units of written Japanese refer to words and syllables, not sub-syllabic elements, like letter-to-sound correspondences.

Even though the JIS computer keyboard in use all over in Japan does incorporate aspects of Japanese literacy, it is still most quickly used for input in computing by those who can touch type using alphabetic units (which are then converted bottom-upwards to syllables and then, when necessary by convention, to Chinese characters). The majority of Japanese can not touch type and find the idea of alphabetically analytic input of language to be alien to their feel and grasp of their own written language in situations that require interaction with a computer or word processor program.

Japanese English learners, especially visually oriented ones, might feel that written English is simply too exotic and strange for them to process visually or analyze into meaningful language. English speakers often have the same reaction to written Japanese. Written English's complex spelling conventions are largely outside the writing practices of Japanese literacy, and, to the extent that those spelling conventions actually reflect a phonetic, phonological or morphological reality, they largely do not reflect anything similar to be found in Japanese.

Complicating this are some of the the linguistic peculiarities of English. In the previous installment, I asserted that it could be argued that English is a Germanic language in terms of its pronunciation, but its spelling makes it look like a Romance language. Well, both of these aspects are foreign to Japanese learners of English. Their language's phonology is not similar to a Germanic one, and their approach to writing and spelling conventions is not a Latin or Romance one.

Connecting with e-mail and to the world wide web only really took off in Japan once G2 and G3 mobile phones became ubiquitous. Little wonder then that the Japanese quickly took to web site addresses that are read like digital bar codes instead of ponderously typed in the standard http:// form.

Also, the interface of Japanese mobile phones is set up for quick syllabic level input and conversion of the content words to kanji. Roman letters are a tertiary system, beneath and much more limited than features for handling kanji and the two syllabaries. However, the linguistic habits of many Japanese have adapted to the need for speed in text messaging. For one thing, this has led to the increased use of roman-lettered acronyms to stand for conventionally shared and frequently used phrases.

Some people have even started complaining because these acronyms are taking on pronounced forms and invading the spoken language. Parents are often shocked when they listen to their teenage sons and daughters conversing in mobile phone acronymic 'short hand' at the dinner table. Such developments are interesting for the ongoing evolution of Japanese as a vital, modern native and national language of international importance in the high tech era. However, these changes will have little or no positive impact on English learning in Japan.

Why English learning fails in Japan: Reason #2

Why English learning fails in Japan: Reason #2

Note: This is the second of a series which is supposed to be planned for at least ten short installments (Ten reasons why...). However, readers' comments are welcomed and will be added as reasons in the main series, with full credit given.

Reason #2: Japanese is not closely related to English

Japanese is not closely related to English.

Language families

What people usually mean by this is that English and Japanese are not from the same language family groups, and that Japanese does not belong to the Indo-European super-family of languages. It is fairly common knowledge among Japanese as well as the foreigners who flock to study Japanese (Japanese as a Foreign Language, JFL) that Japanese is not closely related to any other languages. Actually, what some recent linguistics says is that Japanese and Ryuukuan are members of the larger 'Japonic' family. So technically, modern Japanese (and all its dialects) has a relative, the Ryuukuan language (and all its dialects). At a popular level in Japan, Ryuukuan is often equated with Okinawan, and Okinawan is thought of as one of the many spoken dialects of Japan. The fact that most Okinawans also speak and are literate in standard forms of Japanese only reinforces the popular view, since the same thing could be said about most Japanese. That is, they speak a local and individual variation of a regional dialect, and learn and use standard forms of Japanese for education, business, literacy, etc.

Clearly, we can say English is not closely related to Japonic or Japanese. English is often referred to as a Germanic language. This makes it a sister language of Dutch, Flemish, German, Icelandic, etc. Others might note the close lexical and typographical resemblances modern English has with Romance languages, such as French, Italian and Spanish. You might say that modern English SOUNDS like a Germanic language, but LOOKS like a Romance one because of the Latinate vocabulary and the fossilized French-looking spelling conventions.

Comparing and contrasting Japanese and English

In foreign language education in Japan, English and Japanese often get placed side-by-side for comparison and contrast. This is true of much of the content of EFL classes in Japan, from junior high/middle school to university-level. You could call this the type of linguistic analysis for pedagogical purposes that is thought to support the learning of English, but it is also for the purpose of comparative cultures. In the realm of comparative cultures and cross-cultural learning, reference is most often made to 'American' (but also 'British') English as a standard form for learning EFL. Such an emphasis can be misleading, but it is understandable enough given the relative importance of English as a 'global language' and the global impact of American culture. The fact that the US has a population that is much larger than the UK's, Canada's, Australia's, New Zealand's and Ireland's also deserves consideration.

Unlike modern English, which is demonstrably related to known groups of languages, modern Japanese can be called a language 'isolate' of somewhat obscure origins. Origins of modern languages can be very obscure in the absence of literacy and textual artifacts. And even these have a distorting effect, since we still can not recover fully the spoken language. Such is the case of Japanese. More technically, it should be said that 'Japonic' is the actual family and isolate, and that Japonic consists of Japanese (and all its dialects) and Ryuukyuan (and all its dialects).

English can be classified into relationships with west Germanic languages as well as the larger Germanic grouping. This then allows English to be put into a relationship with a 'super-family' that has penetrated the popular consciousness of language, the Indo-European languages. This means that it can be said that English shares ancestry with German, but also Latin and Greek. However, this also means English shares a line of linguistic descent that stems from an historic super-family of languages that also has lines of descent in Russian and other Slavic languages, as well as languages considered 'exotic' by many Americans or British, such as Albanian, Farsi, Kurdish, Pashtun, Indo-Persian, Hindi and Urdu.

Reinforcing this view that there is a 'Indo-European' nature to English is English's history. English has a long history of language-changing contact with other languages, but the most extensive contacts have been with other Indo-European languages (Celtic, North Germanic, Norman French, literary Latin, etc.)

Like English, Japanese is a 'contact' language. In the case of Japanese, though, the exact nature of the cross-linguistic contact is obscure until literacy practices, a writing system and large amounts of vocabulary were adopted from China. Modern Japanese is most likely a literary creole that is in its origins the results of contact (or waves of contacts) due to human migrations to the archipelago. (Much the same could be said about how modern English was formed.) In the case of proto-Japanese, migrations were followed by cross-cultural consolidations across speakers of N.E. Asian (non-Chinese) languages, S. Asian (non-Chinese) ones, Pacific ones, and ones already spoken on the Japanese islands before the migrations.

Outside of sheer coincidences in typology and traits, Japanese has marked similarities with other NE Asian isolates, like Ainu and Korean--namely, word order (S-O-V), lexical morphology, and phonology. And, like these N.E. Asian isolates, it has been related for the purpose of hypothetical discussion with another broad grouping, namely the Altaic super-family, especially the Tungusic branch of E. Siberia but also the central Asian ones falling under the label 'Mongolian'. This is because just about all the languages being discussed here are typologically speaking, synthetic and agglutinative. However, in terms of its phonology, syllable structure, and morphology, and lexicon, Japanese has also been usefully compared to the super-group of Malayo-Polynesian languages.

Applied Linguistics and Second Language Acquisition

Linguistics existing to support language teaching (LT) and learning (LL) is often referred to as 'Applied Linguistics' (AL). AL is often linked to sub-field called 'Second Language Acquisition'(SLA), even though SLA doesn't contain the sort of content with which linguistics is identified (that is, linguistic description and explanation). The relationship of AL and SLA to LT and LL is attenuated by AL's and SLA's theoretical and conceptual abstraction while its research agenda is often separated from LT by different professional agendas (since applied linguists are based at universities). Perhaps the biggest issue is that most of the research results of academic AL and SLA are not generalizable to larger sets of learners (such as, all EFL students worldwide, or EFL students in E. Asia, or EFL students in Japan, etc.).

However, one common-sense notion that is shared across AL, LT and LL is centered on the relationships of the first/native language (L1) with the second or foreign language (L2, especially when the second or foreign language is not acquired from very early childhood. The idea is simple: If a given L1 (e.g., Japanese) is not historically or genetically related to a given L2 (e.g., English), the acquisition or learning of that language will be more cognitively difficult and take longer.

This idea of language difference in terms of issues in acquisition has been academicized somewhat in AL. AL has at least two ways of realizing the idea in its discourse of theory and research. These are often referred to as 'Contrastive Analysis' (CA) and 'Error Analysis' (EA).

Let us look at Contrastive Analysis first. For CA, the L1's and L2's language systems (e.g., phonology, morphology, lexicon, syntax) are compared and contrasted in detail, and the differences are used to PREDICT issues in learning the L2. For example, in phonology, Japanese is said to have no distinction between an /l/ and an /r/. So CA would predict that Japanese learners of English would have a hard time distinguishing English /l/ and /r/, both in terms of perception of speech and in production. On the other hand, Japanese /r/, at least for an English speaker, actually sounds phonetically similar to English /l/, /r/ and /d/. So /r/ sounds could also be an issue for English speakers learning JFL, marking an accent but actually interfering with intelligibility of their Japanese. Imagine such issues multiplying across differences in terms of phonology, morphology, lexicon and syntax. The implications are complex, to say the least.

Perhaps the issues that arise with CA predicting errors that don't occur while not predicting errors that do is the fault of linguistics and language description. If the language descriptions and concepts you are using to carry out a contrastive analysis are inadequate and faulty, then the errors you predict might be nonsense. They might not reflect any language reality, including the developmental language of learners (often referred to as 'inter-language' in AL. Also, languages are supposed to share some traits almost universally while other traits mark a language as unique. So these marked traits could be difficult for any beginning learner, no matter what the linguistic background. For example, English is an Indo-European language that, perhaps due to significant language contact with other languages, has lost much of its inflectional system. This has huge implications for how tense, mood and aspect are actually achieved when one communicates in English. Therefore, regardless of the language background of the EFL or ESL learner, one can predict that the 'grammar' of the English 'verb' will be a source of a lot of confusion and learners' errors.

Then there is Error Analysis. EA came about because some people noticed that CA predicted errors that didn't occur among L2 learners, but at the same time, L2 learners seemed to experience and produce errors that CA didn't predict. So EA looks in detail at the actual L2 performance of learners and attempts to produce a systematic analysis of their L2 errors and learning difficulties.

Which is not to say that EA is a perfect corrective or supplement to the inadequacies of CA. First, a distinction should be made between random MISTAKES that do not reflect the actual state of the L2 learner's language competence and the ERRORS that do indicate some sort of persistent issue. Second, EA is supposed to generalize to a large group (such as the entire population of Japanese learning English at the beginning level). So the hope would be to detect errors that characterize most if not all of that population. But EA is largely behaviorist in its assumptions, so error behaviors could be linguistically or psychologically misinterpreted by the researcher or classroom teacher using such an approach.

Implications for learning a foreign language

Still, there is a commonly accepted feeling or intuition that, if two languages are not related (such as, they have a different word order or are pronounced very differently), it takes longer to learn the L2. This is not to say that Japanese is so unique in relation to English that Japanese learners of English should have more difficulty than any other language group of learners whose native language is not related to English. There is nothing in the conceptual apparatus of AL or ELT that would support this.

However, since Japanese and English are not closely related at all, this lack of linguistic affinity is something to consider in terms of planning for language teaching and language learning at schools. Unfortunately, it is not addressed adequately. Part of this is the damaging effect that results from Japan drawing on English-speaking countries for the theories, concepts and practices that embody the dominant ideology, technology and infrastructure of LT and LL.

Re-stating the issue

Which brings us back to reason #2 for why English learning fails in Japan. Because of the differences in language typology, the beginning stages of LL will take longer and involve more cross-linguistic and learning issues. These cross-linguistic and cross-discourse issues could range anywhere from phonetics and phonology through vocabulary and syntax and overlap with how to structure a a paragraph, essay, or research report. If we think of learning a FL as a classic 'bootstrap' dilemma, perhaps the core issue here will become clearer. It takes a lot of patience, hard work, repetition, review, and attention to the details of the FL in order for a learner to get beyond the helpless beginner stage. Once beyond this helpless beginner stage, the learner enters a new phase: they will have learned enough of the L2 to succeed at learning it further, perhaps up to mastery, if they are patient enough.

The problem with English in Japan is most Japanese never get beyond the helpless beginning bootstrap phase. They don't learn enough English to learn how to learn it. Unfortunately, if they depend on their schools, universities or the commercial ELT industry to help them, most will not make much progress. For one thing, this is because much of the ELT industry, institutions and programs have failed to acknowledge how and why English is difficult to learn if your native language is Japanese (or, in the case of more and more foreign students, Chinese). They have also failed to appreciate just how difficult it is to teach such a large population of beginning learners. To conclude, this should give you more than a hint of the reasons that will follow (e.g., ELT industry, government policies, institutional management of foreign language education, teacher training, etc. ).