Traditional vs Simplified radicals

[buzhongren]

Ill switch from characters to radicals. Every simplified dictionary I have shows a different number of radicals basically in the 180 range. So that leaves a difference of about 10 radicals plus or minus from one dictionary to another. One dictionary shows 210 and another 175. How come there is no standard like Traditional Kangxi with 214 where any given radical is the same number and meaning. I have more trouble looking up simplified radical index than traditional. The radical is in a different order or may be missing.

xiele,

Jim

[chrix]

moved from this thread

[chrix]

Not all traditional dictionaries follow the Kangxi scheme either. There have been some attempts at reforming radical schemes, regardless of whether we're talking about simplified, traditional or Japanese.

Can you give some examples of characters you can easily find in the Kangxi scheme, but not in simplified dictionaries?

[YuehanHao]

I will complain about this too. Were there a standard, looking up characters would become more efficient. Two paper dictionaries I have use similar ordering (although with a difference of about 5 radicals in total), but the ordering system they use is not the same as I have seen for online dictionaries like YellowBridge or nciku.

In the dictionary I use most often, (without trying) I can remember the radical number for a small number of common radicals (e.g., 心，人，女) and will skip the index lookup step. That's great when it makes sense to use the dictionary I am most familiar with (which is mediocre), but it doesn't carry over to other dictionaries. Similarly I develop a feel for the approximate position of other common radicals in my dictionary's list of radicals of a given number of strokes. Yet when using other dictionaries with different ordering systems, I can't use this knowledge and have to randomly hunt.

Small impact each time, but when thinking of the thousands of characters I have looked up over time, I can't help but be frustrated at this lack of systemization. Other improvements, too, could probably be made if the intent were to minimize lookup times, but having a set standard for the radical index would be a decent first step.

约翰好

[buzhongren]

Chrix:

Every kangxi radical index Ive seen has 214 radicals. There are breakdowns by their number and meaning. The kangxi huo fire radical is 86. Ive never seen a kangxi index with a different number. In one simplified index it is 71(Tuttle Concise) and another 103 (Collins mini). Thats because the four stroke range is big and where they appear is anybodys guess. Most show it in the 70 range somewhere. I have a nice large printout comparing traditional and simplified radicals. Any number I see for simplified is arbitrarily based on the dictionary the printout came from. However the traditional number is always the same and good for any kangxi dictionary. Its just when I recognize a simplified radical whether the same or not over traditional I have to go chasing it down in a simplified index. If I had too I could probably list at least half of the kangxi radicals positions. Even the kangxi yi one radical is a different position in most simplified dictionaries.

YuehanHao: I second your remarks.

xiele,

Jim

[chrix]

That's why I mostly use pinyin-ordered dictionaries, resorting to the radical index only if this fails (usually you can guess using the phonetic component).

I think for a large majority of characters finding them through the radical index is unproblematic, but especially simplex characters (characters that can't easily broken up into components) are hard to find sometimes.

But I feel this discussion would be more fruitful if we could look at concrete examples, we might just be talking past each other here.

To give you one examples of my own, I recently discovered that 裏 裹 belong to 衣 with the radical split in two parts, with the phonetic component between them. Some more modern approaches might actually group them with the "lid" radical.

[chrix]

Every kangxi radical index Ive seen has 214 radicals

Yes, but not every traditional script dictionary uses the 214 radical Kangxi index

[Ah-Bin]

Not all of them use the 214 system, but most of them published outside the PRC do, as do most Japanese, Vietnamese, and Korean dictionaries for kanji/chu nho/hanja, and many Taiwanese (Hokkien) and Hakka dictionaries do too. If you know your Kangxi radicals well, you can use these dictionaries very easily.

Often those that appear to have some system based on stroke number often still use the Kangxi order within the stroke number sections to arrange the characters. (Soothill's Dictionary of Buddhist Terms does this)

and.... Unicode arranges its characters by Kangxi, so when you have to input a rare character you have to go searching for it under the 214 keys.

Years ago (before there were electronic dictionaries) I was encouraged to learn the 214 radicals in order. It took a while, but the result is that I can use most traditional dictionaries just as if they were arranged by pinyin (of course there are some headaches like 相 found under 目 etc. I think it was the best advice I ever had, it helped me learn Japanese "backwards" going directly from the characters to the pronunciation.

So if you are interested in Classical Chinese, dialects, Japanese, or Korean I would advise you to learn the 214 radicals in order too!

[chrix]

this is not necessarily true about Japanese dictionaries. A lot of them have adopted what the Wikipedia calls a 字形 approach that has partially superseded the old Kangxi radicals. This seems also to be the problem with most simplified script dictionaries. (And some mainland dictionaries even seem to have completely abandoned the radical approach and sort characters according to the shape of the first stroke 一｜丿丶乙, but I think this is not what buzhongren is talking about)

But as I've said before, I'd really like to see some examples so we all can see what this means in concrete...

[buzhongren]

Chrix:...but I think this is not what buzhongren is talking about)

The simplified radical index varies from dictionary to dictionary if so present. You would think if simplifying characters was so important why not standard radicals in a standard order like kangxi. Here is one of my guess the radical 尤 characters. Its not dian,yi,pie,ji. Its under a radical by itself almost. However Tuttle Concise does place it under the yi radical. When I learn the phonetics of the HSK characters Ill be able to use a pinyin dictionary without a radical index for 90% of the characters.

xiele,

Jim

[Ah-Bin]

this is not necessarily true about Japanese dictionaries. A lot of them have adopted what the Wikipedia calls a 字形 approach that has partially superseded the old Kangxi radicals.

This is actually true for dictionaries for foreign learners, but the ones Japanese people themselves use usually are Kangxi ordered.

You would think if simplifying characters was so important why not standard radicals in a standard order like kangxi.

Simplifying characters was initially meant to be only a step to the eventual phonetic writing of Chinese, so it shouldn't be surprising that they didn't pay too much attention to radical number and order.

[chrix]

No, Ah-Bin, that's simply not true. There's a number of newer Japanese kanwa jiten that have a modernised set of radicals. It's even mentioned in the Japanese Wikipedia article on radicals, but I also know this from personal experience (my kanwa jiten are all in some box now, or else I'd give you some bibliographical info).

Now one more thing about the Kangxi radicals is that often in the Kangxi dictionary the character meaning takes precedence over the shape, and a lot of Japanese kanwa jiten actually deviate from the Kangxi standard as grouping the characters more according to their shape as their meaning. And modest changes on the number of Kangxi radicals, dropping some underutilised ones..

One thing I like about modern Kanwa jiten is that many of them sort a character at several places, like if a character is known to be miscounted as having 14 strokes where it actually has 15, you can find it at both places, and the same goes for characters that are often mistakenly assumed to belong to a different radical. I find this a user-friendly approach..

Edited March 11, 2010 at 10:30 PM by chrix

[Ah-Bin]

No, Ah-Bin, that's simply not true. There's a number of newer Japanese kanwa jiten that have a modernised set of radicals. It's even mentioned in the Japanese Wikipedia article on radicals, but I also know this from personal experience (my kanwa jiten are all in some box now, or else I'd give you some bibliographical info).

Ah well, I'll just admit to having outdated knowledge then. When you say modernised, do you mean a complete overhaul? Nelson's Japanese English Character Dictionary doesn't stick exactly to Kangxi, but it's almost Kangxi ordered (a few extra bits added here and there).

One thing I like about modern Kanwa jiten is that many of them sort a character at several places, like if a character is known to be miscounted as having 14 strokes where it actually has 15, you can find it at both places, and the same goes for characters that are often mistakenly assumed to belong to a different radical. I find this a user-friendly approach..

Those new dictionary makers are really kind.

[Gharial]

Most simplified dictionaries follow the CASS 189-radical ordering "to the letter" up into the 170s numbers, so it's only up to a dozen or so low-frequency, high-end stroke-number radicals that are going to be missing in any particular simplified dictionary. So what I'd personally recommend is simply making a Kangxi-CASS conversion chart (a sample of a work in progress is supplied in the attached jpeg) for at least those first 170-odd (Kangxi-)CASS radicals (and you could include columns for each separate simplified ditionary you own, with any particular dictionary's missing radicals beyond the 170 mark represented with a dash or N/A), or easier still, buy the ABC C-E Dictionary (original Desk/Pocket edition, not the Comprehensive, which might not have quite the Kangxi-CASS conversion tables anymore - unfortunately I don't yet own and am therefore not familiar enough with the Comprehensive to be sure!)), which has clear and useful Kangxi-CASS conversion information on pages 856-858 (part of Appendix VI) and a less detailed (i.e. not at all visual, just numbers) but still useful CASS-Kangxi conversion chart on page 870 (part of Appendix IX); note however that there are a few mistakes that need correcting in both appendices: http://www.chinese-forums.com/index.php?/topic/23810-errata-for-abc-c-e-dictionary .

[Altair]

Nelson's Japanese English Character Dictionary doesn't stick exactly to Kangxi, but it's almost Kangxi ordered (a few extra bits added here and there)

If that is the same dictionary I used to for Japanese, I want to say that I am still in mourning since I stopped actively studying Japanese. It was darn near perfect. In fact, it was so darn near perfect that I even continued to use it for Chinese for a while, guessing at the origin of the on-yomi to find difficult to locate characters.

If I recall correctly, the dictionary has a series of intuitive rules that let you determine where to find a character almost 100% of the time with only the use of the indices in the margins. 80% of the time they will be under the same place as the original Kangxi rules. In many of the other 20%, there would be cross-references. I remember it being so reliable that I occasionally even used this method over using the alphabetized index, since so many Japanese characters have identical on-yomi.

I started Chinese with 繁体字 and originally learned the Kangxi 214 pretty thoroughly. I then gradually switched to 简体字 and have been frustrated with every dictionary I find. Every one has different radicals, different stroke orders, different indices, different stroke orders for reference, and different principles for ordering characters. I have gradually soured on the whole enterprise and try to use the guess-the-pinyin method almost exclusively.

I cannot recall a specific character that has had a different treatment between the dictionaries, but 羲 is one that recently wanted to look up, because I was unsure how exactly to write it. I can't remember what dictionary I used, but I decided that beginning with 羊 was all too likely to lead to disappointment, and so I remember that it was the same as the right element in the traditional form of 牺, looked that up, saw the traditional equivalent, and finally had my character.

There must be a simpler way!