Outlier Linguistic Solutions

[OneEye]

OK, I think it's a good time for me to start an official thread here. I've posted a bit about Outlier elsewhere on the forum but I'd like to have a dedicated thread for it here.

 

Note: We've just launched our Kickstarter. Check it out!

 

Outlier Linguistic Solutions is a recently-founded publishing company in Taipei. Our primary project right now is to create a dictionary of Chinese characters aimed at learners. Our goal in this is to take the latest research in Chinese palaeography, phonology, and etymology and repackage it into a system that's easy to use for zero-level beginners in Chinese who have no training (or even interest!) in those fields.

 

We explain characters in terms of their functional components (sound and meaning components), and explain exactly how each component functions within a given character. That is, we'll explain whether a given meaning component (大, for instance) express meaning by way of its meaning (big) or by way of its form (a person). We also have a system for explaining sound components, which I can't talk about just yet (journal article forthcoming, after which I can). Our system not only explains how characters work in a clear, easy to understand way, it also helps with long-term recall and the ability to predict the possible meanings and pronunciations of new characters that you encounter.

 

The dictionary will cover the 3000 most common characters (in both simplified and traditional), with an additional volume planned later with another 3000 for those who require more. Both an English and a German edition will be published initially, with hopefully other languages to follow, and every edition will be entirely bilingual (both Chinese and the base language). Each head word will have two sections: a basic, "this is what you need to know to master this character" section, and a "for experts" section (what I like to call the "red pill section") for those who want to dig into the etymology more deeply. We also plan to suggest mnemonic objects for each component that people can use to create mnemonic stories.

 

They asked me to join as a research consultant (I'm now also in charge of marketing), which I was thrilled to do because I really believe that this project will be a huge help to learners of Chinese. We're designing exactly the kind of tool I wish I could have had access to when I started learning Chinese. I used Heisig, which has some really great things about it, but could be much stronger if it retained its focus on mnemonic stories but also explained characters in an etymologically sound way and didn't ignore phonetic components. That's what we're making. I also used Harbaugh's book (aka zhongwen.com), which is also really great — I especially like that it covers simplified and traditional characters, mainland and Taiwan pronunciations, and has tons of vocabulary words — but it has a lot of flawed, outdated etymology. Our dictionary will have all those features but will make use of the latest research in etymology. We'll also be able to continue to update it as new advances in the field are made, because it will be released in a digital format (with a print edition planned for later).

 

Ash, our tech director and lead researcher, has published journal articles and book chapters and presented papers at major conferences about historical phonology, palaeography, and Chinese character pedagogy. He's currently co-authoring a paper with Dr. Feng Sheng-li, the author of Expressions of Written Chinese. He's given a talk at ICLP about how to learn characters every year (sometimes twice a year) for the past 6 or 7 years. This fall, if enough students sign up, he'll be teaching a class there on the subject. That's all to say, he really knows his stuff when it comes to Chinese characters, not just from an academic standpoint but also from a pedagogical one. The dictionary has been in the works for several years now. I've actually had former ICLP students complain to me that Ash had mentioned it in his talk ~5 years ago "and it still isn't done!"

 

We're making plans for other projects besides this one, which I'll announce when the time comes. But for now, I wanted to get the word out. We have a website and a blog, where I've recently posted two articles: one I co-wrote with Ash on the etymology of 高興 and one on 六書, the traditional way of categorizing characters and whether it's useful for a learner. More articles coming soon.

 

As excited as I am about this, I'm going to try to avoid bombarding the forums with every minor update because I know that can get annoying. So if you want get updates about new blog posts, videos, conference talks, etc., we're also on Facebook and Twitter, so check us out there.

 

So there it is. I'm here to answer any questions, hear suggestions, listen to publishing proposals, take abuse, what-have-you.

[MPhillips]

Sounds like it will have some of the virtues of Karlgren's "Analytic Dict. of Chinese & Sino-Japanese" (characters grouped by shared phonetic component; etymologies) minus its defects.

[OneEye]

Well, I don't know if we can hope to be favorably compared with Karlgren (though it's a nice thought). But we can definitely present a picture of the Chinese language that's a lot more accurate. Karlgren did great work with the materials he had access to, but there have been major advanced since his time and his stuff is now very out of date.

 

One of the cool things about doing this digitally is that we have a lot of flexibility in how we present things. In a print dictionary, grouping characters by phonetic component (or 部首, or pinyin) is nice, but in a digital format it doesn't have to be either/or.

 

This is how we're conceiving it: let's say you're looking at a basic sound-meaning character like 粒. 立 is the sound component, and 米 is the meaning component. You could click on 立 and it would show you the entry for that character as well as:

• All the characters in which it's a sound component (like 粒 or 位) and
• All the characters in which it's a meaning component (like 站)
• All the characters in which its current appearance is a corruption of an earlier form that was not 立 (like 音 or 童)

That last point is important, by the way, because it tells you that 立 in a character like 音 has nothing to do with the sound or meaning of the character, that it's a placeholder for something else*.

 

Another fundamental difference to Karlgren's book is that ours is aimed at learners of Chinese, most of whom probably have no background or interest in palaeography or linguistics. I talk a lot about etymology here because I know that a lot of the users here have sufficient knowledge and interest to be able to relate. But in our dictionary we're going to strive not to overburden the learner with etymology for its own sake. The etymological research is in the background, which we then distill into just what the learner needs to know. The minimum effective dose, if you will. We will have a section that presents the etymology in more detail, but it isn't necessary in order to use our system. It's there because we love this stuff and we know others will too. It's also there so that, for instance, Chinese teachers and professors who want to be able to recommend something with a solid base in academic research to their students can see that our dictionary offers just that (the one who comes to mind most readily is Victor Mair, who has railed against many of the popular books for learning characters on his Language Log blog).

 

But yes, we are hoping that we can use the good ideas that other authors have had in the past and also incorporate new research and sound pedagogical principles. I don't know of any other book for learners that even comes close to getting etymology right. The problem is that the ones who have the skills and expertise to get it right aren't generally interested in helping non-native learners. They're too busy reading excavated texts and such. The ones who are interested in helping learners don't have the expertise in palaeography or linguistics, so their work always falls short on that front. Two of us are trained in palaeography and linguistics, and Ash spent six years in the PhD program in Teaching Chinese as a Second Language. So we have the training on both sides to be able to do this the right way. Also, all three of us are non-native speakers with professional-level proficiency in Chinese, so we know what it's like to be a learner and what students will look for in a resource like this. I think that's a good combo (though I'm biased), and hopefully it will turn out to be everything we want it to be.

 

_____________________

* Red pill section: In this case (see attached image), 音 was originally just 言 with a mark put there to distinguish the two characters — a dot inside the 口. 言 in turn was originally a 舌 with a distinguishing mark at the top. This is a great example of why thinking of characters in terms of 部首/radicals is error-prone. These three characters are all 部首, so it would be tempting to think of them as unrelated, but they're quite closely related to each other both in meaning and in form. It's just that they've evolved in different directions since their creation.

[MPhillips]

I agree with your philosophy--many legitimate etymologies are very memorable--it's not always necessary to resort to fanciful Heisig-style mnemonics in order to aid the learner.

[OneEye]

Right, and another, very fundamental problem with Heisig-style approaches is that they ignore the role that sound plays in writing, essentially treating every character like a semantic compound (會意字). Writing has three components: form, sound, and meaning. Ignoring any of those is detrimental. Chinese characters (form) have sound components and meaning components. Learning characters via their functional components makes it possible to use the sound and meaning of a word to recall the form (that is, how to write the character). If you know the character you want to write is pronounced jiāng and the meaning is a river, it's easy to use your knowledge of sound components (what can sound like jiāng? 將、工、畺 etc.) and meaning components (what's likely to be in a character that means river? 水) to recall how to write the character (江, of course).

 

The point of all this (one of the points, anyway) is to strengthen the learner's ability to recall any character they've learned with a high degree of accuracy over a long period of time. Mnemonics fade away over time (though they can be very useful in getting the character into your head in the first place), but a systematic knowledge of how functional components work will serve you well for a very long time.

 

Another benefit to this sort of system is predictive ability. If you know what pronunciations 尚 (for instance) can represent as a phonetic component, then when you run into 淌 for the first time, you'll have a good idea of how it might be pronounced and what it might mean (especially if it's in context).

 

A lot of people ask us why we place so much importance on being etymologically accurate — teachers have told us that their students use method X and get good grades on their tests. Those two points (long-term recall and predictive ability) are the answer. We're not aiming to help people ace their dictation tests (which is a very short-sighted metric), we're aiming to help people get literate in Chinese, to be able to use it as a tool over the long term.

[wibr]

Awesome project, I also used Heisig and would love to see a more comprehensive approach. What will the digital format be? A simple pdf/epub, integration with pleco or your own website/app?

[anonymoose]

Sounds like what zhongwen.com has.

[xiaokaka]

Sounds like an awesome project! Integration with pleco would be the way to go!

[Lanchong]

Keep up the good work. There's so much regurgitated stuff on the Chinese-learning blogosphere that it's easy to think one has seen it all. But that article on 高興 was a breath of fresh air.

[OneEye]

What will the digital format be? A simple pdf/epub, integration with pleco or your own website/app?

 

TBA. ;) We don't have anything locked down just yet, but I'll announce here when we do.

 

 

 

Sounds like what zhongwen.com has.

 

I talked a bit about zhongwen.com in the first post. His etymologies are flawed and very outdated because he based his book/site almost entirely on the 說文解字 (one noted palaeographer has said, and I'm paraphrasing, that it isn't outrageous at all to say that the 說文解字 is wrong most of the time. It is nearly 2000 years old, after all.). Ours will be based on the latest research, plus it will be in a much more usable format, not to mention a host of other advantages over his book and others. It's probably the best thing out there currently, but after all, Harbaugh is an economist and not a specialist in etymology.

 

Another point where ours differs with Harbaugh's is that we will have semantic charts showing a basic outline of how the meaning has changed over time. So you'll learn how a character got from its original meaning (亦 meant armpits, for example) to its current day meaning. The only meaning that has a direct relationship to a character's form is its original meaning, so for characters like 亦, this is especially helpful.

 

 

Thanks for the kind words, everyone! I should mention that we're also open to suggestions for content ideas. The post on 高興 was suggested by one of our followers on Twitter, for example. If you have a question about etymology, studying Chinese characters, or anything else that we think merits a post or video, we'll do one!

[Angelina]

Not just the blogosphere. People actually bought into the Chineasy thing because there couldn't find anything better. Finally something professional coming.

[OneEye]

Well, to ShaoLan's credit, she (or her team) has some serious marketing skills. I'd love it if we could get a TED talk, write-ups in The Wall Street Journal, and all the other media attention she got, but I wouldn't even know how to start going about it. From a design standpoint, too, they did a great job. But that's where it stops, unfortunately. Her character explanations are folk etymologies at best and made-up gibberish at worst, she doesn't even talk about pronunciation, and her "qualification" for doing the book is that her father is a calligrapher. Well, mine is a chiropractor. Anybody care for a spinal adjustment?

[Yang Chuanzhang]

I'm not sure Chineasy was successful simply because there wasn't anything better. Some people just seem to want to have their belief confirmed that all Chinese characters are pictograms...

 

Anyway, really excited about this project! Will the entries in the dictionary be similar to your blogpost on 高興? And if not, is there any chance you could give us a teaser of what the final format will look like?

 

Also, is it possible you could give any estimate of when the dictionary will be done?

[OneEye]

Will the entries in the dictionary be similar to your blogpost on 高興? And if not, is there any chance you could give us a teaser of what the final format will look like?

 

Well, we want to avoid overloading people with information. That kind of thing will be in the "red pill" section, but the info will be more condensed than in the blog post.

 

As far as what it will look like, we're working on that. It will, of course, depend on the platform via which we release it. Once we get that finalized, we should be able to post a mock-up of what the final product will look like. I can give you a general timeline:

 

1. Ash is working on getting his paper on sound components finished and getting a promise of publication from a journal. That has to happen before we can talk to anyone about our approach to sound components. I even had to sign an NDA for him to tell me about it. He's a little protective of it. This might take a month or two, depending on the journals he shops it to.

2. After that, we'll hopefully be able to nail down the distribution platform. I'm not sure how long this takes, but I wouldn't imagine it's a really long process.

3. Then we can show you the mock-ups.

 

Also, is it possible you could give any estimate of when the dictionary will be done?

 

We're hoping to finish within a year or so. To continue the above timeline:

 

4. Once that's done, I have to go back to Taipei for a few days so we can shoot our pitch video...

5. ...for our Kickstarter. We've put a ton of our own money into this, but the research itself involves hiring people to do data entry, create and maintain our database, library maintenance, not to mention office rent and upkeep, etc. We're keeping our costs as low as possible (and they probably get sick of hearing me ask, "And how much is that? Are you sure we need it?") but it still gets expensive. We could pitch the project to investors, but we don't want to lose creative control over the project to a 行外漢 if we don't have to. On that note, we're open for suggestions about possible perks for different donation levels.

6. After that, it's just (just!) a matter of finishing the slog through all the research, writing the thing, and getting it published. But we're writing it in three languages (more accurately, we're writing it in Chinese and then translating it to English and German), so it takes time. The upside of this is that releasing it digitally should significantly cut down on the amount of time it takes.

 

During the whole process, we'll continue putting up blog posts and videos about Chinese characters and how to learn them, as well as other topics like how to improve your tones, etc. We've also got a list of blogs that we're hoping to guest post on (one of which is already confirmed), so look out for that, too.

 

By the way, our blog is now here. It should be accessible from China now, but if not, let me know. Chinese and German versions are coming soon.

[xiaokaka]

The blog is working for me now, however the images aren't loading unless I click on them. Don't know if this is related to the GFW or something else.

[OneEye]

OK. For right now, we're just redirecting our Wordpress.com blog to our webpage, so that may be the cause of it (I think the images are still hosted on Wordpress). We're working on migrating the whole thing over though, so hopefully that will fix it.

[mouse]

A really interesting project, can't wait for it to be ready.

 

As for donation perks, it's tricky. Apart from the obvious branded detritus, I would say access to exclusive content would most interest people, but that would mean more work on your side or risk keeping potential promotional material like the blog posts you've already published hidden. Maybe early access to information would encourage people to donate, something like showing research notes, little titbits of info that you weren't able to use. Cutting room floor stuff. Maybe if you made the "red pill" section subscription only, then gave donators free access; six months for x amount, a year for a bit more, something like that. Of course, I'd rather it all be free, but you have to pay for all this somehow. Oh and please don't actually call it the "red pill" section in the finished product. That would be embarrassing.

 

PS Wordpress images etc all have the wordpress domain name, so will automatically be blocked in PRC. Host them somewhere else and you'll be fine.

[OneEye]

Don't worry, the red pill thing is just a joke. I'm not sure if I like the idea of making that content subscription-only or otherwise cost extra, and it would certainly make things more complicated.

 

We were thinking about doing things like exclusive videos or ebooks for donors at a certain level. For instance, the content of Ash's annual ICLP talk would make for a good video, and it's something the vast majority of people can't get access to normally. We're also thinking about offering a live webinar or even 1-on-1 Skype consultations. Hadn't considered branded detritus, though. Outlier coffee mug, anyone?

 

But yeah, maybe research notes or other behind-the-scenes stuff would be an option.

 

Thanks for the ideas.

[li3wei1]

I think that certain blog posts or dictionary entries, perhaps condensed and with a bit of design magic, would be appealing on a t-shirt or coffee mug. What better way to say "I'm studying a really cool and difficult language, and I'm pretentious enough to tell everyone about it"? I'd buy the mug. The question is, which character? A set of 5000 would be impractical.

[wibr]

- Custom calligraphy if you know some calligrapher who wants to support your project as a top perk
- Postcards/shirts ... with fancy oracle bone characters
- Sponsorship for individual characters, so that somewhere my name is listed as sponsor of “龜"
- Limited prints: Special gold edition, cheat-sheets (most common components, phonetic components, ...)
- I also like the idea of getting additional videos