I’m 15, it’s the summer of 2007, and I’ve just arrived in Japan for a 6-week homestay program made possible by my stupidly awesome high school. At the end of the first week I get an email from my sensei back in America asking me how I’m doing, etc.
“Everything’s wonderful,” I reply, “but why do Japanese people talk so fast?!”
Eventually I got used to it – right around the time my summer ended.
But the question sat quietly in the back of my mind ever since. I chalked some possible answers up on the board:
- the syllable structure flows with lots of short-length vowels, making it possible to speak a thousand words a minute. English words don’t let you go that fast. For your own safety.
- the words themselves are just so long. Speak any slower and you’d forget what you were trying to say by the time you got to the middle.
- related to the above – the words might be long and full of syllables, but given context it’s easy for native speakers to fill in the blank once they hear the start of the word, letting the speaker move much faster without overwhelming the recipient with information.
- and finally, because maybe that’s just the way it is.
I never looked any deeper in to the question.
It’s worth the read. Because it turns out Japanese people talk 27% faster than English speakers.
Koichi has some cool theorizing on the reasons behind some of this in the article I linked. Something he brought up that I didn’t realize: Japanese has just 416 possible syllables, compared to 1,191 in Mandarin, and 7,931 in English.
My most important takeaway from the study: Japanese has the lowest “information density” of the eight studied languages. I.e. you have to use more syllables to convey the same amount of information.
As you might expect, Japanese speakers spoke the fastest – used the most syllables-per-second – but despite that fact, the language still manages to convey ~75% of the actual information conveyed by all eight other languages.
Perhaps this suggests there is simply a physical limit on syllabic rate and Japanese sits at that threshold. Or maybe the limit is on listening comprehension – 15 year old me would certainly agree with that. Either way, it’s seems that other languages allow a greater rate of information transfer – in fact, every other language transfers information between speakers at essentially the same rate. Except Japanese.
So to increase Japanese’s “information rate,” you could either speak faster or use fewer syllables to convey the same points. Speaking faster, well. My untested armchair hypothesis is that speaking faster is impossible and that Japanese already exists at the limit of rapid speech.
Use fewer syllables? Well, perhaps this is where some outrageous (to foreigners) Japanese contractions come in. To some degree, I think all languages do this – contract and manipulate their language (when speaking with the rest of your “tribe”) until the “information rate” is maximized.
Will not to won’t.
たべました to たべた
But even then, Japanese is fundamentally constrained by its available syllables. With only 416 combinations, mostly consonant-vowel pairs like “KA,” “KI,” “KU,” “KE,” “KO,” the language struggles to abbreviate and shorten words without mis-communicating entirely.
Japanese is missing many of the consonant-vowel-consonant options available to English – something that I found often caused my fellow high school students in Japan, who at the time were studying English, to struggle. “Sing”, for example, is not a valid construction in Japanese. You can’t just end with a “G” like that. Nor is my name for that matter. Conrad was written and pronounced “Con-ra-do.” A commenter gave the English example: “wolves.”
So here’s where it gets interesting. And even more hypothetical.
If you ignore Kanji for a moment and imagine written Japanese across all forms of media – newspapers, magazines, books, essays, signposts, ads, and more – I think the situation would be quite similar. Could we find another language that manages to write so much, yet convey so little?
I don’t know. But I think it would be tough. And it wouldn’t be English.
Why is this interesting?
Because written Japanese is actually incredibly information-dense. Reading the newspaper for non-native speakers is an exercise in painstaking translation and spot-the-hiragana.
Kanji. A writing tool appropriated from Chinese with its 1,000+ syllables and normal rates of “information transfer,” and then grafted onto Japanese.
One small character suddenly replaces two, three, or more characters in our hiragana-only script. At its heart, I think this is why I found studying kanji so difficult. I could speak the language, understand what was being said, and I even had a script to write it all down in. But it wasn’t enough.
The “information density” contained in a writing tool like kanji has to make up for the abysmal density of pure hiragana. And it does. When you can achieve some measure of kanji-literacy, reading becomes a breeze. Ambiguity disappears. Paragraphs shorten and become faster to read simply because your eyes don't have to travel so far. But until you’ve achieved that mystical level…damn. It’s hard to stop feeling angry at whoever decided to appropriate the Chinese script.
Of course, writing Japanese without kanji is like writing English without….what? What other language has a reasonable analogue? I’ve not been able to find one. Not in our Latin-derived languages. Not in Chinese. Not in Korean. Where else can you write down everything you want to say, and then go back and replace the majority of characters or letters with a symbol that represents groups of individual characters?
It would be like replacing the word “burger” with the McDonalds logo, that big curvy yellow “M.”
Except of course sometimes the big curvy “M” means cloud, or yellow, or sand dune. So sometimes you see it and say “burger,” and other times you see it and read aloud “sand dune.” It breaks the link between visual and audio learning, because you have to learn all three readings, even if you hear one every day and the others only once a month. And so once a month you find yourself reading something that doesn’t make sense, stop, pause, go back, think think think-
Ah. Of course. “The camel crossed the sand dune.”
Facetious, I know, but not wholly inaccurate.
And I think that’s what makes kanji so difficult for non-native speakers: a combination of depth (3,000+ to memorize), inconsistent visual and audio patterns (multiple readings), and sheer information density. But those same traits also make it worth studying.
 Note that the study’s sample size is neither particularly large nor particularly small. I’m taking their findings as the literal truth here instead of debating the study’s demerits. Given the surprisingly consistent data for all languages except Japanese, I think – regardless of the degree – it’s reasonable to conclude that Japanese is materially different.
Meet Kanjimasu. I wrote last time about fixing my broken Japanese language skills, but I didn’t explain how I plan to do it. Lofty goals without grounded plans aren’t worth much.
There aren’t a lot of resources out there for budding Japanese learners like me. But they do exist. So why am I building my own?
Partly because everything I found was either expensive or designed with last decade’s best-practices in mind.
Now I’m not saying I’m a great designer. Far from it. This site is my third time designing anything at all, and I take no credit for the good parts. But as a user and a learner, I can’t deal with janky UI elements, complex navigation, tiny graphics, no mobile functionality, and all the other quirks that come with free software.
Which is fine. Free is free is free. Is another man’s pride and joy.
But I need more. And I think I can do better.
And the other reason has nothing to do with Japanese, learning Japanese, or studying for the JLPT: I discovered coding.
Not for the first time – I read a book on Objective C 2.0 in high school because I wanted to know how those new iPhone apps were being made. I took a class at MIT on Java because I’d enjoyed that book and wanted to actually make something without declaring Computer Science.
But despite it all, I never really built anything. Never believed I could do it without a degree and a large team. Never tried to. Until I took MIT’s 6.470, taught myself Ruby, Rails, JQuery, and HTML/CSS in four weeks, and launched Short Story Board.
That first project wasn’t very good. But when I stumbled into my problem with JLPT kanji I knew I could fix it. What’s more, in addition to learning Japanese I’d be improving my coding skills. If we’re talking about expanding our personal knowledge – and I often am – it’s the perfect solution: learn something new, use it to build something I want, learn even more!
So here I am. Introducing Kanjimasu. If you’d like to do so, feel free to use it to study kanji and their compound vocab words. I will be.
 Codeacademy, Medium, Hacker News & others were my sources of inspiration. Perhaps you think things are getting worse, not better, (I don’t. Most days.) but the general rule of thumb so far has been “Less is More.”
 Coding, hacking, developing, programming, software engineering, computer science…there’s a scale in there somewhere. As an initiate with grand ambitions I don’t want to overstate my abilities lest I offend those more senior.
 Stupid untested belief. Maybe a hold over from Mechanical Engineering. Maybe just a lack of personal need/will.
 Of course “taught myself” is a bit bold. I picked up enough to cobble together a poorly performing, embarrassingly ugly Rails app in four weeks. But I loved it. Designing and building things rocks. I need more of it.
 感ます. Cross-lingual puns are rarely successful in both languages. English-speaking students of Japanese might get it. To the rest, I apologize.
Recently I’ve been entertaining the idea of moving to Japan for 9 months, maybe a year, sometime in the not-so-distant future. I’ll get into the why of that decision later. It's just a pipe dream at this point. But I studied Japanese in high school and for my first two years of college, I’ve been to the country 3 times, and I once spent a summer there. And now I’m a graduate...and two years is a long time to go without much practice.
Current language status:
Vocab – mediocre to begin with. Casual watching of Japanese TV and anime has helped keep this up somewhat, but I don’t feel comfortable saying I could understand a conversation anymore.
Listening – see above.
Speaking – probably my biggest disappointment. I used to be comfortable opening my mouth and letting the words flow forth in a mostly-intelligible stream of middle-schooler Japanese. But now? I freeze up mentally, half-formed sentences dying in my throat. My declining ability brings my confidence down and the less confident I am, the less I practice, the worse I get, the less-
You get the idea. Shy death spiral.
Writing – objectively my weakest area. But honestly I don’t care about this so much right now. It’s not that I don’t understand how important it is to be able to communicate with the written word...I’d just like to know the words themselves first. Statistically also the activity I'm likely to use the least.
Reading – my reading was fastest and most comprehensive 6 years ago when I spent that summer in Japan. 6 years ago, despite studying the language through 2011. When I was actually there being able to read was like a mini-superpower. I could go places, read maps and signs, follow directions, find the one store that still had a Nintendo DSLite in stock…
You get better at things by doing them, language or not, and while I was in Japan I had to read in order to get what I wanted (and speak too, to a lesser degree). So I did a lot of reading and speaking, and not much writing.
And while I still hear Japanese on a semi-regular basis (TV shows), I never see it written in the course of my average day. Predictably, my reading skills suck.
[Quick aside for non-Japanese speakers – Japanese has 3 modes of writing. Hiragana (48 simple characters each corresponding to a syllabic sound, used for native words), Katakana (another 48 simple characters corresponding to the same set of sounds, used to write foreign words), and Kanji (logographic Chinese characters according to Wikipedia. The complex little pictures non-comprehending foreigners tattoo on their body because someone told them it meant Love. Or Dragon. There are 50,000 of these little bastards).]
After a recent reevaluation I realized I’d actually forgotten some katakana. Ouch. Speaking I could get around by virtue of never doing it. Watching subtitled anime was enough to deceive myself about my listening skills. But forgetting how to read kana?
In principle, in practice, in any use-case, it’s the same as forgetting how to read the letter “Z”. Sure, it might not be used that much. But you don’t forget the alphabet.
So at this point I have a choice. I can accept that I studied a language for over 6 years, spent multiple months in the country, watch(ed) native TV shows, and at the end of it all…have nothing to show for it beyond the memories. “Once I knew Japanese.”
On the other hand, I could dive back into it, spend the (hopefully brief) time getting back to where I used to be, then build on that knowledge base even more. “I know Japanese.”
There’s arguments for both sides. Realistically I have much better job/life prospects in the US. Japan is no longer an ascendant nation either, so it’s not like I’m putting myself between an important (read: growing) trade relationship. Honestly, if there’s any economic gain to be had from learning Japanese…I can’t see it.
A little mercenary, I know, but if I’m going to invest in myself, spend the time studying and learning, it makes a lot of sense to invest in something that has a meaningful chance of having a return. And it makes sense to cut the waste.
On the other hand, I believe in the future. Wishy-washy, I know, but here’s the way I see it: at any moment in time the set of knowledge and skills I have should be larger than any time before then. Not just different, but larger. Extend that forward into the uncertain future and I get progression. I find the idea of holding steady, static, deeply saddening – an acknowledgement that things have already peaked. And regression? Unthinkable.
But I can’t hide from it any more. My Japanese has regressed massively. The only thing that could excuse me would be if I learned enough about other subjects to not only offset my declining language skills, but to lift me far above where I used to be. Well? Has it?
I don’t know. I got a degree in (Mechanical) Engineering over that time. But something small I learned in that process was that I’m not cut out to be a mechanical engineer (story for a later date). So did I trade one set of “useless” knowledge, Japanese, for another?
I can’t answer for sure, of course. But I do know that it doesn’t feel like personal progression. Things feel very static.
So what am I going to do about it?
I’m going to fix it.
Or at least try my damned hardest to fix it, while working 100 hour weeks at my day job.
Goal: Pass the Japanese Language Proficiency Test (JLPT) Level 2