蟒蛇 uwabami——a drunken python

In English, if we want to show off a little while writing, we can mine the thesaurus for obscure words (at risk of obnubilating our message and discombobulating the reader). With tens or hundreds of thousands of words to choose from, the vocabulary of English represents a practically unlimited resource to play with. But what if, rather than words, there were 10s of thousands of letters or characters, the majority of which were unfamiliar to even the best-educated reader? What if I could write something to you in your own language, and you have no idea what it says because you have never seen those particular ᖃᓂᐅᔮᖅᐸᐃᑦ before? In English, the concept seems absurd because we only use 52 characters and a few punctuation marks to write the language. If we come across a new word, we may not understand it, but we can at least read it. But it’s quite different in Japanese.

One morning in the office after a work night out, a colleague got an email from the section manager, describing him as 蟒蛇. Having never come across this word, he had no idea either how it was pronounced or what it meant. Some kind of snake? Why was the buchou calling him a snake? His fellow team members gathered around his desk, but they didn’t know either. Looking it up online, they discovered that it was pronounced uwabami and that it refers to a giant python or a great drinker (a reference to the previous evening’s activities, which had involved rather a lot of beer). It seems that there is an old story about a giant snake that ate and drank a great deal. What was interesting for me is that the first character, 蟒, is used only to write that one obscure word and in no other context in Japanese.

As it happens, the section manager makes a study of obscure kanji and has demonstrated a very high level of knowledge of the Japanese writing system by passing the “pre-1” level of the kanji kentei exam, an exceptionally rare achievement of which he is rightly proud. (Passing Level 2 is considered evidence of a good university education, while level 1, the most difficult level, is only for the most dedicated kanji scholars, with only a few hundred obsessives passing it every year.) Success at pre-1 level means knowing thousands of kanji, both their common and unusual readings as well as unusual uses in compound words and old Japanese proverbs. So it’s not so surprising, having amassed such a wealth of learning, that you might want to share some of it with your hungover colleagues in a light-hearted e-mail on a Friday morning. Even if it means showing off a little!


やさしい日本語のニュース—Easy Japanese News


I often watch the Japanese TV news but I have a lot of difficulty in understanding it. The vocabulary is too advanced and the sentences too complicated and too quick for me to follow.

NHK offers a service for learners like me: it’s called News Web Easy. Every weekday, three or four stories from the day’s news are presented in simplified vocabulary, read slowly and clearly by a woman in an easy-to-understand voice. There’s a good variety of interesting topics. All of the kanji are glossed with furigana, and you can click on underlined words to see a definition.


After you have listened to and read through the easy version, you may then feel brave enough to click on 普通のニュース futsuu no nyuusu—the “normal” news clip. Of course this is much easier to understand once you know exactly what the story is about.

It’s interesting to compare the two (the easy and the normal versions) and see what changes are made. A big part of it consists of replacing “Chinese” or Sino-Japanese vocabulary with native Japanese vocabulary. For example, instead of 変化 henka, they might use the Japanese verb 変わる kawaru to mean “change”. Similarly, if they want to say that someone has been arrested, instead of using the word 逮捕 taiho, they substitute 捕まる tsukamaru.

This confirms something from my own experience: that native Japanese words are easier to learn, easier to remember and easier to understand than their Sino-Japanese equivalents. But why should this be? After all, the Japanese words are usually longer and more “complicated” looking than the Chinese loan words. Strangely, I think that this is a big part of the reason.

Typical Sino-Japanese vocabulary is bisyllabic, with words consisting of two simple syllables (drawn from a restricted set of possible syllables). This very simplicity has the effect of making the words less distinctive and memorable, and leads to lots of homophones. For example, 成功 seikou means success, but there are numerous other words, such as 製鋼 精工 性交, with the same pronunciation. Was the newsreader referring to success, steelworks, Seiko watches, sexual intercourse or something else? Usually the listener is just expected to figure it out from context.

There is another way, however: Japanese TV routinely captions its programs with Japanese subtitles. Not just news, but all sorts of general entertainment – if someone on screen is saying something, you can read it on screen. So even if you are not very proficient at understanding spoken or written Japanese, the fact that both are available can be a big help.


I recommend News Web Easy to learners at about level N4 or N3 who struggle with understanding normal Japanese news broadcasts, and would be interested in following current events while learning lots of good topical vocabulary.

Why learn bushu? Building-blocks of kanji

In your first year of learning Japanese, you probably learned the days of the week:

  • 月曜日   Monday
  • 火曜日   Tuesday
  • 水曜日   Wednesday
  • and so on.

And, with one exception, you will also have learned all the characters needed to be able to write these words; simple, Grade 1 kanji such as 日 sun, 月 moon, 火 fire, and 水 water. But the fly in the ointment is that tricky 18-stroke character 曜 you in the middle of each of the weekday names. Until you somehow get to grips with that, you may be able to read the Japanese word for Monday, but you won’t be able to write it.

Grade 1 kanji © Dara Connolly
Grade 1 kanji © Dara Connolly

Learning to write simple kanji is a straightforward matter of repetitively copying out the character until it is imprinted in your mind and muscle memory. Which is fine for the first 80, or maybe even 200. But over time you will have discovered that these simple characters are not typical, and that this method doesn’t scale well.

Fortunately, as you encounter more complex kanji, you will have noticed that many share features in common, and that those features suggest the meaning or sound. For example, the characters 痛 painful, 病 sick, 疲 tired, and 疾 shame all have the following feature in common: 疒. This “radical” or bushu*, known as やまいだれ yamai-dare, gives you a clue that the character means something to do with sickness. Other common bushu include radical 149 訁gon-ben, which appears on the left side of characters to do with speaking, radical 140 艹 kusa-kanmuri, which sits on top of kanji that refer to plants, herbs or vegetation, and radical 85 氵sanzui,  which suggests a watery meaning.

There are 214 classical bushu in total, and during the past year I invested the time to learn each of them. Even the obscure, archaic and seemingly useless ones. Not just to recognise them, but (importantly) to write them. So was it worth it?

Well, let’s consider the character 曜 you, mentioned above. No longer is it a scary, seemingly arbitrary assemblage of 18 strokes; it is now a simple-to-remember construct of just 4 parts: 日 + 彐 + 彐 + 隹.

Each of those parts has a nickname, so I can describe it in words: nichi-hen, kei-gashira, kei-gashira, furutori. (sun, pig’s head, pig’s head, old bird. And remember that I’ve learned to write each of these parts, so it’s trivial to put them together and write the character. A character that, like so many others, I have been able to recognise for 20 years but would not previously have been able to write.

On the other hand:

  • The bushu were not actually designed or selected for this purpose. They are not a comprehensive list of kanji building-blocks, nor were they ever intended to be. Their purpose is to be dictionary headings, to allow you to look up characters in the dictionary. Some very common building-blocks are not bushu, but are still worth learning. For example, 寺, which appears in characters like 詩 time, 待 wait, 持 carry, 侍 samurai, 特 special, and so on, usually giving a sound clue (“ji”) rather than a meaning clue. You really need to know those too.
  • The list of bushu is not “efficient”; many of the bushu are themselves made up of simpler bushu. For example radical 186 香 ”fragrant” is made up of radical 115 禾 ”two-branch tree” and radical 72 日 ”sun”. This is okay though, as the more complex ones are often kanji worth remembering in their own right.
  • Some of the simplest radicals are just lines or dots, with limited semantic content.
  • Some of the bushu are utterly obscure. For example you will almost certainly never meet any kanji containing radical 35 夊 sui-nyou “go slowly” or radical 191 鬥  tatakai-gamae “war”. But I’m a bit of a completist, so I learned them anyway. And in the case of radical 192 鬯 nioi-zake “sacrificial wine”, it paid off (see below).

But the day I knew it was worth it was when I noticed that I could now write the following famously-difficult character:

utsu – depression

From being an absurdly complex and impenetrable mass of lines, it now resolves itself as just six parts to remember, each of which I already know how to write: 木 缶 木 冖 鬯 彡(tree, can, tree, cover, sacrificial wine, hair).

So why learn bushu? Four reasons:

  1. They’ll give you clues to the meaning of the kanji
  2. They’ll give you a short cut to remembering new kanji, without having to explicitly learn them
  3. They’ll make your knowledge of kanji more precise (for example, allowing you to clearly distinguish similar kanji)
  4. You need them to be able to look up characters in an old-fashioned dictionary (does anyone do that anymore?)

And maybe a 5th reason, if you’re anything like me:

5. It’s kind of interesting in its own right and gives you more insight into the writing system.



Is it worth learning the bushu during your first few years of learning Japanese? I’d say almost certainly not. You can put your time to far better and more enjoyable use learning vocabulary and grammar, listening, speaking and reading the language. But later—maybe much later—you may come to feel, as I did, that your knowledge of the kanji is built on a somewhat shaky foundation; that you can recognise many kanji in context but you don’t really know them; that you still get confused between kanji such as 通 and 進, or 速い and 遠い (or worse, you never realised that 着る kiru—to wear and 着く tsuku—to arrive are the same kanji!) And you’ll want to go back and consolidate your knowledge, deepen your understanding. And when that time comes, you could do worse than set aside some study time to learn the bushu.



* The word 部首 bushu literally means “section head”; each radical sits at the head of a section of the dictionary, and that’s how you look up the character in the dictionary. They are known as the “Kang Xi” radicals after a dictionary of 47,000 characters published in 1716 on the orders of Chinese emperor Kang Xi, and they still form the basis of paper dictionaries in Japan and China to this day.




カルタ—playing cards

I received a present yesterday evening.

We met my former Japanese teacher and her family for dinner, in what has become an annual tradition, at the Turkish/Mediterranean restaurant アルピーノ Alpino. It was an evening of delicious food and convivial company, as well as an unbeatable Japanese learning experience to be surrounded by Japanese conversation, including the excited chatter of a 7-year-old girl who sat beside me and chatted away to me all evening.

The present came neatly wrapped in a yellow bag from Osaka souvenir shop ichibirian, the bag decorated with typical images of Osaka and some phrases in Osaka dialect. Opening it up, I found what I thought was a book, with the title “Osaka PhD”.

Front of card game “Osaka hakase”

But instead of opening like a book, the middle slid out to reveal two sets of playing cards; one with text and the other with colourful images of fun facts about various Osaka neighbourhoods.

Osaka waiwai karuta

The “front cover” (or maybe back cover, from a European point of view) says 大阪わいわいカルタ oosaka waiwai karuta—Osaka clamorous cards, and then in smaller writing underneath, “Did you know? Didn’t you know? This and that about Osaka”. The whole cover is brightly decorated with typical Osaka images such as Osaka Castle, Tsutenkaku tower, bunraku theatre, fugu, and a bowl of ramen.

Osaka waiwai cards: や

Each card pair is associated with one hiragana symbol. For example, the text card for や (ya) has the following “interesting fact”:

八尾市はな 歯ブラシ生産 日本一

yao-shi wa na, haburashi seisan nihon ichi

Yao city is number one in Japan for toothbrush production.

The corresponding image card shows a woman holding a toothbrush atop a smiling Mount Fuji, symbolising the number one status.

The text on the cards uses little furigana symbols alongside the kanji for the benefit of children (and me!) for whom the task of learning to read Japanese is still a work in progress.

I was really struck by the ease and fluency with which musume-san read the text on the cards. I expected a child of that age to be painstakingly sounding out the words, but she just read them off at full speed (much more quickly than I could). I wondered if she was just familiar with the text of the cards from playing with them at home, so I tested her by showing her some sentences on my phone that she had never seen before. She showed equal facility reading the following random sentence:


doraibaa de neji wo mawashita ga, nakanaka umaku mawattekurenai.

I turned the screw with the screwdriver, but it just didn’t want to turn.

天才娘さん tensai musume-san

I guess she is some kind of a prodigy; I assume most Japanese 7-year-olds don’t read at that level. I asked her mother about it and she just replied that “she loves reading”.

Sensei, danna-san and musume-san after dinner with me and Yuko in Alpino

Anyway, it was a lovely evening and we all left feeling feeling happy and saying また来年 mata rainen—see you again next year!

だらだら dara dara—idly

My name doesn’t have good connotations in Japanese. In one local dialect on the Sea of Japan coast, dara means “idiot”. (Perhaps fortunately, I have yet to visit that area.) It also features in the word 堕落 daraku, which refers to a moral lapse or descent into apostasy, corruption, sin or depravity. And in the word darake which refers to being completely covered in something (generally something bad, like mud or blood), or filled with mistakes.

And then there is this phrase dara dara, meaning slovenly, idly, slowly, lazily.


This biscuit tin features a very popular character called “Rilakkuma” (relax bear) looking characteristically relaxed, with the slogan (written in Roman letters) “kyou mo minna de daradara goron”, which means something like “today also, everyone idly idle about”. I bought the biscuits because I felt the word dara dara was being used in a nice, positive context.


dara dara is one of hundreds of gitaigo, so-called mimetic or onomatapoeic words in Japanese. For example:

  • They were seated bara bara (separately);
  • The stars were shining kira kira (glittering and sparkling);
  • She was laughing kusu kusu (giggling);
  • She was laughing gera gera (loudly and boisterously);
  • She was laughing hera hera (condescendingly);
  • He speaks English pera pera (fluently);
  • Rain can fall zutsu zutsu, shito shito, pota pota, potsu potsu, depending on the intensity.

This aspect of Japanese is really hard for the learner. Japanese is simply filled with these words: people sleep guu guu, they eat mogu mogu, they lick pero pero, crunch food gari gari, stare jiro jiro, get nervous doki doki or impatient ira ira

Even with flash cards and other learning aids, they just seem to defy memorisation. Part of the reason must be that, despite being known in English as onomatapoeia, they are mostly not in any meaningful sense mimetic, but rather seemingly arbitrary. Another possible reason is that these words somehow don’t “feel like” real, proper Japanese, but like some kind of add-on, possibly childish or slangy. (For example they are always written in kana, not kanji.) This feeling is incorrect; they absolutely are an integral part of the language, including the literary language, but it’s hard to shake it off.

Together with another type of adverb (of the form bikkuri, yukkuri, shittori, pittari, kussuri, ukkari…), countless hours of effort are spent trying to memorise these vocabulary items for the JLPT exams. Effort which is mostly wasted, since this kind of knowledge (lists of arbitrary items learned by rote memorisation) is only shallowly rooted in memory and is quickly forgotten once the exam is over.

It’s different, however, when you learn one of these words in “real life”; somehow hearing it used even once in the context of a conversation anchors it in reality and instantly makes it much more memorable. And once you use it yourself, it’s with you for life.

語彙 goi—vocabulary

When I was travelling to work on the subway in Japan, I often saw high school students studying on the train, on their way to school. And when I peeked over their shoulder, they always seemed to be studying the same thing: English vocabulary. Specifically, English verbs.

The vocab book had about 10 or 12 verbs on each page. Each entry had the definition in Japanese, and an example sentence in English. The students made use of a sheet of ruby-coloured translucent plastic which they would move down the page as they read, uncovering the verbs one at a time.

A number of things struck me.

First, that the vocabulary they were studying seemed extremely advanced (and extremely random). A typical page might include words like (and I am making this up because I can’t remember specific examples):

vindicate, overpower, intervene, culminate, becalm, entreat…

Second, that it seemed like an extraordinarily ineffective way of learning a language. Learning long lists of words and their meanings is very difficult, and does not really equip you to use the words in context. And I know this because…

Third, this is exactly how I studied Japanese for my JLPT exams. My Japanese vocabulary books were an exact mirror image of the English vocab books the Japanese high school students were studying on the train. A Japanese word, the English definition, and one or more sample sentences; 10 per page. Week after week, month after month, I would try to cram 10 new vocabulary items every night. But they never became “real” for me until I later read them or heard them in the context of real-life spoken or written Japanese. The vocabulary I was learning was far in advance of my actual ability to speak Japanese, and was focused entirely on the requirements of the exams.

The same was true of the grammar books; 4 “grammar points” per day that I had never encountered in real life but needed to know for the exam.

Was it useful? Certainly, it was not a waste of time. It built up a very solid foundation of passive theoretical knowledge which could later be “activated” as my “real” level of ability, lagging far behind, caught up to the point where I might understand or use some word or grammar point that I had memorised for the exam. Each subsequent encounter, in context, reinforces the knowledge.

Was it the best way to learn Japanese? I would say not. There is a place for rote learning, but it needs to be balanced with exposure to the language as it is actually used. After the exam in December, I put away the books and changed my focus to a more natural way of learning, which included private one-to-one conversation lessons as well as reading Japanese texts. This proved very effective. But part of the reason it was effective was that I had already learned a lot of vocabulary and grammar.

One thing I would warn well-meaning teachers (and writers of textbooks). If you tell a student “here are two similar words; make sure you don’t confuse them”, the one thing you can guarantee is that the student will be hopelessly confused between those words for years, if not for life.

For example, if a student comes across the words amido (a screen door, for keeping out mosquitoes) and amado (a storm door or shutter) for the first time, on the same page of the vocabulary book, they will find it very difficult ever to remember which is which unless they can come up with some trick or mnemonic to distinguish them.

Other examples:

  • gurasu means a glass (a drinking vessel) while garasu means glass (the material). I’m pretty sure (but not 100% sure) I got that the right way around this time!
  • airon means an iron (for ironing clothes), while aiyan means iron (the material). For years I could never remember which was which, and could not use either with confidence. Only recently, when the film Iron Man (AiyanMan) came out, was I able to remember.
  • In Lithuanian, šaltas means “cold” and šiltas means “warm”. Good luck remembering which is which. (It helps if you think that  šaltas is cognate with German kalt.)

But if you were to first encounter those words separately, each in context, and learn them in that way, you would never confuse them.

Note on the word of the day:

The Japanese government maintains a list of “kanji for everyday use” that are supposed to be used in newspapers and so on. Following some recent modifications in 2010, the list now consists of 2,136 kanji characters.

In reality many well-known and fairly common kanji are not on the list. Most notably, the list includes relatively few of the characters for types of animals, fish, trees and other living things; some of which characters even I would be familiar with.

The word 語彙 goi meaning vocabulary includes a fairly obscure character 彙 which was not on the list until it was added in 2010. It was one of a batch of 196 characters that were added to the list at that time, while a few lesser-used characters were removed (a character for “spindle”, for example).

One of the characters that was added received a lot of media attention at the time: 鬱 utsu, meaning gloom or depression, a character of great complexity, written with 29 strokes.


逆 gyaku—backwards

Jay Rubin (the translator of many of the novels of Haruki Murakami into English) wrote an entertaining little book called Making Sense of Japanese: What the Textbooks Don’t Tell You.

It’s a book I’ve come back to again and again over the years, as it has something to offer for learners at different stages. And it’s amusingly written; Dr Rubin has a comic gift.

One of his memorable chapter titles was “Warning: this language works backwards.” His point being that many Japanese sentences can most easily be understood by starting at the end and working backwards to the start.

This is especially true for sentences that are moderately complex. For example, sentences that contain a relative clause.

To test the theory, here’s an example of a moderately complex sentence from this week’s homework (an article about robots):

Original sentence: これは手足の不自由な人が家庭内で自立するのを助けるために開発された。

Breaking this down, we have:

  • これは these
  • 手足の of limbs (arms and legs)
  • 不自由な without full use (disabled)
  • 人が people
  • 家庭内で within the home
  • 自立 independent
  • する be
  • のを助ける help to
  • ために in order to
  • 開発された were developed

Reading from top to bottom (left to right, in the original sentence), we have the following incomprehensible gibberish:

these of arms and legs without full use people within the home independent be help to in order to were developed“.

However, starting at the end of the sentence and working backwards, things start to make a lot more sense:

were developed in order to help to be independent within the home people without full use of arms and legs these“.

If we take the word “these” (the topic of the sentence) and move it back to its rightful place at the start, we now have an English sentence that is both grammatical and meaningful:

these were developed in order to help to be independent within the home people without full use of arms and legs“.

Basically, the more complex the sentence, the better the “read backwards” trick works (I use it all the time when I am trying to puzzle out the meaning of long sentences). If a sentence ends からだ ”it’s because”, that’s a good starting point for me to understand the sentence.

There are two main points where the trick breaks down:

  1. In Japanese, subjects and topics are normally found at the head of the sentence, just as they are in English. English is a Subject Verb Object (SVO) language: “Brian picked berries”; whereas Japanese is a Subject Object Verb (SOV) language: “Brian berries picked”. So in both languages, the subject, “Brian”, comes first, but the order of verb and object is reversed.
  2. In both English and Japanese (unlike, say, French or Spanish), the adjective precedes the noun: “red berries” rather than “berries red”.

And so, for simple sentences like “Brian picks red berries”, the trick doesn’t work.

All this raises the question: do Japanese people have to wait until the end of the sentence before they can begin to understand it by working their way through it backwards? I presume not. They must be able to understand their own language in a forward direction.

For me, however, as a learner, the sentence doesn’t really start to take shape until I’ve heard the verb (at the end of the sentence), and a noun phrase remains adrift until I hear the noun (at the end of the noun phrase).

I encounter many “garden path” sentences in Japanese, where what I take to be the subject and verb of a sentence turn out, as the sentence continues, to be part of a relative clause modifying the real subject of the sentence. Such sentences “lead you up the garden path” and either leave you confused or force you to change gear halfway (like the English sentence “The complex houses married soldiers” or “Fat people eat accumulates”).

Will I ever be able to understand Japanese “forwards”? I guess I will have to wait and see.

Just for fun, here’s another sentence from the article:


  • また furthermore
  • 人間 human
  • や and (and so on)
  • 小型の of small-size
  • 動物の of animals
  • 形 the form
  • にしていて made in
  • 簡単な会話 simple conversations
  • ができる can understand
  • ロボット robots
  • も also
  • ある there are

Translation “Furthermore, there are also robots that can understand simple conversations, in the form of humans and small animals and so on.

(Google Translate: “In addition, there is also a robot you are in the form of small animal and human, can be a simple conversation.“)