Outlier Linguistic Solutions

[vellocet]

You need to hype up the fact that the outdated radical concept has been dumped once and for all.  That's huge.  That's something that I've been impatiently waiting to happen for years now.  That's something I didn't know until I read the link above. 

[OneEye]

It is huge, but we're finding it difficult to sell that idea to people. For a lot of native-speaking teachers, it's blasphemy. For a lot of people who don't know the difference, radical = component, so they hear "don't think of radicals, think of radicals!" But for the people who get it, it's a big deal. We just need to work on a simple way to explain it (building it up like I did in my article takes much too long) so that more people get it.

 

In other news, a piece about us just came out in Tech in Asia. The journalist, Josh Horwitz, did a piece on Mike Love and Pleco last year as well.

[studychinese]

OK, supported with $50 in kickstarter.

[tiantian]

I would like to know how well the whole "components" concept works with simplified characters and how much work went into character explanations for simplified characters.

 

We all know that lots of the simplifications used are kind of inconsistent, with inconsistent switching of radicals compared to the more structured traditional forms. There are several simplified characters where the original etymology and meaning kind of got lost in simplification.

 

Since you are Taiwan based you are probably focusing on Traditional Chinese. You say you support Simplified Chinese as well. If your integration of Simplified Chinese, however, just goes as far as switching the font, many explanations will probably render pretty useless.

 

On a screenshot on your Kickstarter page it says something like "To understand this character look at the traditional form". Since I am not interested in learning the traditional form, this is kind of pointless in my opinion.

I fear that this dictionary will be of limited use for many simplified characters.

[Shelley]

My big question is where have you got your resources from? What I mean is has there been some new information come to light? Or is this merely a rehashing of existing resources and information?

 

The reason for these questions is I am trying to work out how you have arrived at your explanations.

 

As far as I know there is a finite amount of resources when it comes to the history of characters.

 

I understand how over the years explanations and meanings have become corrupt for various reasons, are you, as it were, untangling all these mistakes and misunderstandings?

 

How can I be sure your explanations have not got the same type of mistakes that have crept into modern characters.

 

I hope these questions are taken in the spirit in which they are asked, that is one of genuine curiosity and interest. I am not trying to undermine or criticise your work which of course I have been unable to actually read and use yet.

[OneEye]

@tiantian

 

Great questions.

 

I can assure you we're not treating simplified characters as inferior in any way. We're fully aware that most people are learning simplified, not traditional, and we're treating them with equal care. The "empty component" concept is much more important with simplified even than it is with traditional, and it's already very important there. Empty components don't express a sound or meaning, they're just a form, and that's the case with a lot of the cursive-type simplifications like 監/监.

 

There's much more that goes into it than just switching the font. In fact, you can see an example of how we're handling the traditional/simplified issue by comparing the entries for 監/监 and 謝/谢 in the demo. The latter is very simple, simply switching 訁 for 讠, and doesn't require much to be said about it. But there's a much more complex answer to the question "why does 监 look this way?" so the explanation is a bit more complicated. That info doesn't have to be memorized, of course, it's just there to show that there's a good, logical reason for two vertical lines to be there (and really, there's a good, logical reason for most simplifications even if it isn't obvious from the surface-level forms). It's enough to know that the two vertical lines are an empty component.

 

We only explain simplified character in terms of their traditional counterparts when it's necessary, which isn't very often. 监 was chosen on purpose as a demonstration of one of those instances. We're not out to convert anyone to traditional characters.

 

As a side note, when we showed David Moser the demo, the 監/监 entry was one of the things he was most excited about, because he said there are some simplified characters that still don't make much sense to him.

[OneEye]

@Shelley

 

No worries at all. They're great questions.

 

Ash and I have quite a bit of academic training doing this type of research (he has much more than I do), and we're consulting the work of the best, most reliable scholars in palaeography, historical phonology, morphology, and other fields (a few off the top of my head: 季旭昇、陳劍、裘錫圭、劉釗、黃德寬、陳偉、杜忠誥、鄭張尚芳、William Baxter 、Laurent Sagart 、Edward Shaughnessy). There has been a tremendous amount of new material brought to light over the past century or so, especially since the 70s, both palaeographic and phonological/morphological. Our understanding of how Chinese characters work has been completely overturned. As recent a scholar as Karlgren has made the assertion that there's no need to look at anything earlier than 小篆, which we now know to be completely inadequate. There was a great deal of corruption before 小篆, for one thing, and for another you really have to look much earlier to understand sound representation in Chinese characters. The vast majority of these new findings are known only to specialists in these fields, which is why we say we're "distilling ivory tower knowledge" and presenting a lot of info in English for the first time.

 

So we're basing our explanations on the work of these scholars and others, though as trained scholars ourselves we're not past disagreeing with them when we feel the evidence doesn't hold up (that's rare, of course). Those instances would be marked, naturally. The framework for explaining the characters is our own, based on a paper Ash presented a few years ago at a conference in Korea (attached to this post above). We came up with our own framework because existing frameworks (and every scholar has his or her own pet system of categorization) that allow for correct explanations are intended for specialists, not for learners. The paper is titled "A Learning-centered Framework..." to reflect that ours is intended for pedagogical purposes, not for palaeographers. The sound formulas which will be in the official release are also ours — it's a project Ash has been working on for a while and he'll be publishing his findings once the research is finished.

 

We don't really regard the graphic evolution as being incorrect per se, because we're not assigning a value judgement. The term "corruption" causes some confusion here, but we're not talking about some pure, good platonic ideal original form versus an inferior version. We're simply describing what happened historically. However, if you're interested there's a bit more nuanced discussion in the comments here.

 

There are certainly things that can't be known for sure given the presently available data, and we're careful to mark those things when we can. Similarly, our understanding of some things may change as new evidence is brought to light. We do plan to update the dictionary when that happens — releasing it electronically is great for that — but we can say that our dictionary will, as far as is possible given the fact that we're human, present the most up-to-date information available when it's released.

[studychinese]

Who else on the forum is supporting/buying on Kickstarter? We should help to make this thing happen.

[OneEye]

The Tech in Asia piece is now on Mashable.

 

And I've already posted this link, but I have to say, John Pasden did a better job pitching our product than we've been able to. I'm learning. 

[Guest realmayo]

Who else on the forum is supporting/buying on Kickstarter?

 

It would be a fun toy to have but I can't persuade myself that I'd actually find it useful.

 

You need to hype up the fact that the outdated radical concept has been dumped once and for all. 

 

I don't see the problem with radicals. Sure they're slightly artificial. They served a need (paper dictionaries). And still serve a need, of providing a simple (artificial, temporary) support to help learning characters: "this part is meaning, this part is sound".

 

If you uproot (!) the radicals, are they replaced by an equally simple framework to support the learner?

 

Because from a learner's point of view having to remember that in this character " 大 looks like 'big' but is actually a beautiful headdress", but in that character "大 is a corrupted form of a character which no longer exists" etc etc ... will make memorising characters much harder?

[Shelley]

In the Any thing with....thread you mention

the sheer volume of stuff being dug up (literally — excavated Warring States-era bamboo strips have proven to be some of the most important materials for this stuff)

 

This is the sort of thing I wondered about, is there actually fresh data to draw from?

And it would appear so.

 

I wonder if a new system will be more helpful to new learners rather than established students who will have forget and relearn things.

 

Also I suppose that unless all dictionaries, lexicons etc use your system, one will have to learn two methods of classifying and systems of looking up in dictionaries.

 

Don't get me wrong I am really into the idea of a radical new way of organising and accessing characters I just hope it doesn't throw the baby out with the bath water and simply add confusion rather than help.

 

I am now going to sit back and wait for the actual dictionary to be able to make further comments about it.

[OneEye]

@realmayo

 

I think you're misunderstanding a bit, but it's likely due to unclear explanation on our part. Well, maybe. I never said 大 looks like "big" but is a headdress (you're thinking of 羊). But it's not nearly as confusing as you're making it out to be. It's actually perfectly in line with the three properties of Chinese characters:

 

All characters have 1) a form, 2) a meaning, and 3) a sound. For example, the 1) form of 大 is "a person," the 2) meaning is "big," and the 3) sound is dà.

 

1) When it's depicting a person, as it does in 美 or 天, it's a form component.

2) When it means "big," as in 尖, it's a meaning component.

3) When it's representing sound, as in 达, it's a sound component.

4) If it's doing none of those things, it's an empty component.

 

I think that's quite simple and easy to grasp.

 

But if you get rid of two of those categories, you muddy the water because now you have to make up a story for what 天 has to do with "big," or what a sheep has to do with "beautiful." That may be fine and memorable on the level of an individual character, but it obscures the real connections between characters.

 

One of our core ideas is that the more clearly you understand how the writing system works, the easier it is to learn new characters. There are four types of components. That's perhaps a bit more difficult to grasp than two, but it elucidates what's happening on a systemic level so in our view, it's worth the very minor extra mental burden. You don't have to know what an empty component used to be. All you need to know is that it's not representing sound or meaning, and it isn't depicting something. The explanation of how the 羊 in 美 got that way (it's standing in for a headdress) is just so you know there's a logical reason why it isn't functioning in one of those ways. You don't have to memorize that part unless it helps you to remember the character. It's usually enough just to know that there's a reason, so your brain can accept it instead of wondering "what the hell does a big sheep have to do with beauty?"

[OneEye]

@Shelley

 

Also I suppose that unless all dictionaries, lexicons etc use your system, one will have to learn two methods of classifying and systems of looking up in dictionaries.

 

We're not proposing a new way of looking up characters in dictionaries. We're proposing ditching the pedagogical concept that characters are composed of radicals, because it's inaccurate. Radicals are used to look up characters in dictionaries, and that is their only use and function. That's precisely what the 部首 system was designed for. The hint is in the name: 部 section 首 head. A radical is simply the head character of a given section of a dictionary. The word "radical" in English is a very poor translation — they are not roots/radices — but we're pretty much stuck with it now.

[Shelley]

Ok I think I understand a bit more but I will wait so I can comment with some idea of what I am talking about

 

I will try to reread what you have done so far with what you have said in #133 in mind.

[Guest realmayo]

One of our core ideas is that the more clearly you understand how the writing system works, the easier it is to learn new characters. 

 

Yep I think this is where I'm naturally sceptical. Slightly exaggerated but no one would advise beginner students of English to remember what words are from Latin, which Greek, etc (even though personally I really like all that stuff).

 

Perhaps it is the difference between memorising and understanding. To memorise lots of characters, it is clear that no deep understanding is necessary. Just look at Heisig. Or Chinese people, come to that. And once you've memorised a character, you don't need to understand its history, just its usage.

 

But if you were learning characters in a slow steady way, maybe 2 or 3 a day every day for three years, then you've got more time to use the deeper understanding of a character that this dictionary offers, as a memory aid.

 

I do think that the current radical system has its use because some kind of system is needed to help break down and remember characters. If the more 'correct' and more detailed version you are selling places no greater load on the mind that the current alternatives (e.g. radicals, mnemonics) -- then no harm done!

[Guest realmayo]

All characters have 1) a form, 2) a meaning, and 3) a sound.

 

I genuinely don't understand what would you mean by the "form" of, say, 暮 ?

[studychinese]

OneEye, thanks for this. I am actually fairly proficient, relatively, at reading characters. That said I would like a more systematic method of acquiring new characters, and evaluating known characters.

 

My pet hate of character learning is the mnemonic method, the attachment of ludicrous stories to characters to assist in memorization. Some people swear by that method, but it has never worked for me (I am one of the many learners for whom mnemonics do not work at all). To the extent that this project could kill the red herrings afflicting Chinese language learning, I think that this is a great thing.

[Altair]

Who else on the forum is supporting/buying on Kickstarter? We should help to make this thing happen.

 

I have not yet done so, but intend to within the next few days.  I look forward especially to the "expert" edition.

 

I have actually fantasized about creating just such a product.  I am a serious language hobbyist, but still just a hobbyist, and so have neither the proper credentials or time available to do such a thing.

 

I do not think I have close to the typical background for the prospective buyers, but one of the aspects that attracted me were some of the explanations that traced the final shape of the characters through alterations between cursive and regular script.  I peeked at the explanation for 临 and was pleased to see this covered.  I was first exposed to such information in reading an explanation for the final form of 应, which I had not realized had specific cursive origins, and always wanted more such information.

 

My first full exposure to a method like this was in reading A Guide to Remembering Japanese Characters by Kenneth G. Henshall.  He explained the best Japanese scholarly understanding of the origin of the 2000 basic characters and also provided a short mnemonic for each one.  For the longest time, this was my go-to book, even for Chinese characters.

 

One thing that the Outlier project might benefit from is more explicit commentary on mnemonic issues.  Many users will conflate what this project does and what methods like Heisig's actually cover.  For instance, the Outlier project should explain why it might actually matter to the average learner that the top part of 美 is a corruption.  Many learners may feel that this is irrelevant to their mnemonic system and so is useless clutter.  They may wonder why Outlier does not seem actually to have any mnemonic system at all.

 

According to my view, which I think is consistent with Outlier's view, what seems like useless clutter in the short term is usually quite helpful in the long term as the complete aspects of how Chinese characters work is increasingly understood.  In cases, like 美, knowing that the top is a corruption helps explain why there are no sound or meaning correspondences between the modern character and the two apparent components.  Knowing that the bottom is just a front view of a person helps explain that the notion of "big" is not really relevant here. Knowing both these facts also helps explain the physical arrangement of the components, which does not seem to match a meaning of "big sheep/goat."  Armed with this knowledge, nothing stops you from ultimately using "big sheep/goat" as a mnemonic or learning whatever the "official" radical/陪首 is, but you will now be aware of the sound issues and the possibility that both 羊 and 大 (and similar characters) may play similar roles in other characters.

 

Another thing that might benefit from more explicit discussion is a clearer explanation of how a lot of background information can be conceptually and mnemonically summarized and actually lead to quick recall.  For instance, in recalling what 监 represents, what appears to be a long explanation can be summarized for mnemonic purposes as something like:

 

"What remains of an inspecting eye and a bending body over a reflection in a bowl of water".  This was meant to depict: "inspect" or "oversee."

 

By choosing the summary words well, you can help the learner key into important associations, such as the reduction of 臣 ("inspecting eye") to the remaining 刂, the potential link between the reflection and the fifth stroke, and the link between "bending" and "oversee."  For learners like me, this is a much better mnemonic technique than working with the surface meaning or shape of components like "hanging knife 刂," "half of bamboo 竹," and "bowl" and trying to relate those meanings to "inspect" and "oversee."  It is also better than being left with only the etymological explanation without a "hook" to help recall the import parts.

[Altair]

 

Quote

All characters have 1) a form, 2) a meaning, and 3) a sound.

 

I genuinely don't understand what would you mean by the "form" of, say, 暮 ?

 

I also think this could be more simply explained, but I think the idea is as follows.

 

The inventors of characters used a certain (1) shape or form to suggest a visual image.  That image was often linked to a particular spoken word with a (2)  meaning and a (3) sound.  As characters developed, sometimes a particular shape or form was used not for any particular association it might otherwise have on its own with a particular spoken word, but because it helped depict a larger scene in which that image could fit.  In 美, although there is an idea of a person, with or without outstretched arms, there is no idea of "bigness"  and no association with the specific word dà.  Thus, it is used only for its form.

 

 

1) When it's depicting a person, as it does in 美 or 天, it's a form component.

While I accept this explanation for 美, I look forward to hearing why it is true of 天.  I have always assumed that this depicts a person holding outstretched arms to the vast sky above and that in this case "bigness" was highly relevant.

[Kobo-Daishi]

Congratulations on reaching 10%!

 

Well, we've passed the 20% mark!

 

I was hoping there was a way to track a breakdown of how many people are donating per day and their amount, but, there doesn't seem to be any.

 

While looking around, I did find this funny article about a potato salad project that got funded to the tune of $55,492 (when they only wanted 10 bucks) and it has a breakdown of their donations across the life of their project.

 

https://www.kickstarter.com/blog/potato-salad-by-the-numbers

 

Amazing!!!!

 

I really wonder what was going through the minds of the people who donated after the $10 goal had already been met.  

 

Let's just give this guy money?

 

Kobo.