Outlier Dictionary of Chinese Characters

Actually that turned out to be not China's fault at all but rather Amazon's - our file distribution system is backed by Amazon S3 which went down yesterday taking like half of the internet with it :)

I'm still getting constant fails on Archon - which I've not had issues with before
 
Are there manual downloads available for:
1174275316-p2outlierexpertsc-170210.zip
2066525324-p2outlierexperttc-170210.zip

Pleco keeps asking me to go to pleco.com/androidfiles but I'm not seeing Outlier on there.
 

mikelove

皇帝
Staff member
Really, even now? We're on a new China download server that should be much more reliable - are you by any chance on a phone from a Chinese OEM that was recently updated to Android 7? Which brand / model?

We're reworking that page as we try to deal with this National-People's-Congress-related server mess (turns out it wasn't just Amazon) - generally you can get at any file by prepending http://d26r4pfa6wznjb.cloudfront.net/ to it, including those. (skip the initial numbers plus -, just start at the 'p2')
 

Miguel

状元
Working fine here in China both on Archon and Android (used vpn to download, don't know what would happen without since I haven't tried it)
 
I'm on Archon too. (Wondering if Archon isn't passing it's data through my VPN). I see the files get made on my machine and then they fail and delete.

edit:

btw - I'm on 12VPN - I had tried the following settings: WEB, WEB + Mail, WEB + Socks and nothing was working - openVPN wouldn't work at all - just now I was able to connect to StealthVPN (Singapore) just now and I can get Pleco working again.

Seems like ARChon doesn't work over the WEB, SOCKS or MAIL.
 
Last edited:

rizen suha

状元
Hi.

Below 4 examples of characters with (some of) their respective meanings and (some of) the meanings of their components (marked with *).

In all 4 cases, both components would seem to carry meaning related to that of the character. In the outlier dictionary none of the 4 characters have more than one "semantic" component.

The 4 characters were picked from a randomized list of 274 (HSK4) as the first 4 that seemed to be good examples. These 4 characters were found within the first 11 characters.

I believe to be seeing a similar pattern in around 40% of the 1200 characters that I know – and in at least 20%, the pattern would seem to be as clear as in these examples; with no need for opportunistically choosing particular (and perhaps marginal) meanings of characters and components to complete the puzzle.

So, I personally do not feel this to be a case of "pareidolia" (seeing teddy bears in the clouds), but I may be wrong.

Any thoughts?
And (perhaps related to the above discussion, perhaps not so)...
Will the outlier dictionary include "meanings" of components that maybe historically were not considered to be semantic components?
Will it include discussions of these components – and if a component historically was not conceived as a semantic component, maybe in the course of time its very presence has tainted the semantic orientation of the host character?

I am very much a beginner, so forgive me if I am posing a lot of silly questions and conjectures.

Thanks

-------------------

篇 essay, paper
* bamboo
* flat, tablet

忽 neglect, unawares
* must not, without
* heart, mind

符 tally, conform to
* bamboo
* deliver, entrust

漫 flood, all over
* water
* long, extended
 
In all 4 cases, both components would seem to carry meaning related to that of the character. In the outlier dictionary none of the 4 characters have more than one "semantic" component.

I passed this on to Ash, and after looking at it, here's his response:

Super short summary: It is very likely that the sound components in these characters do not have a meaning: 篇, 忽, and 符
This one is inconclusive: 漫

This is a very interesting question as it brings up a lot of issues. One of the main things is that the modern meaning of a character or component isn't what's important in deciding whether a given component has meaning or not. Rather, you have to look at the original meaning of the character and the meanings that a given component may or may not have had when that character was invented. Please note that I'm not saying we don't make mistakes. We do, and when we find them, we correct them.

Another issue is that many characters get re-interpreted as being meaning-meaning (會意字) by later generations. This is due to confusing form components as meaning components. Most meaning-meaning characters are very pictorial in nature like 祭 jì “sacrifice” which depicts a hand (又) placing a piece of meat (top left) on an altar (示). This is very diferent in nature to very late characters like 尖 jiān “sharp” (literally “small over large”), which rely on the meaning of the components (rather than what they are a “picture” of). To make sure we don't fall into this trap, we need to see sufficient evidence before saying that the sound component also gives a meaning.

I'm going to go over your examples in a very brief way (at least, I'm trying to be brief). If you're interested in seeing the evidence I've used, just ask. I'm not including it now as it would make the post really long.

Ex1: 篇 essay, paper
* bamboo
* flat, tablet

篇 originally meant “ancient writings on bamboo strips tied together with a piece of thin rope or leather strap.”
扁 originally meant “large characters written on a sign or placard and posted somewhere.” 冊 depicted bamboo strips tied together and 戶 is a door (half of 門).

Analysis:
The original meanings of 冊 cè and 篇 are very close. Both are writings on bamboo strips that are tied together. The original meaning of 扁, though it's related to writing, is still a bit different. It did not have the meaning “flat,” though it could mean “essay,” but not the kind that is written on paper. None of the sources I looked at showed it as having a meaning.

Ex2: 符 fú tally, conform to
* bamboo
* deliver, entrust

Analysis:
“tally, conform to” is not the original meaning of 符, but rather an extended meaning. 付 originally meant “to give something to someone.” The original meaning of 符 was to match together a broken off piece of bamboo to the other original half of that piece of bamboo. This was done in order to verify that orders delivered to someone where indeed from the proper authority (and not from anyone else). The meaning “to conform to” is an extension (by way of generalization) of this meaning. The meaning “to give something to someone” doesn't really match, since the point of 符 is the verification process, not the giving of the bamboo piece. None of the sources I looked at (and I looked at quite a few – I can show a list if you're interested) showed 付 as having a meaning, including the《說文》.

Ex. 3: 忽 neglect, unawares
* must not, without
* heart, mind

Analysis:
勿's meaning “must not” or “not” is by sound loan, though this meaning appears quite early and was very likely around at the time 忽 was invented. If you can find examples of 勿心 in an ancient text (prior to the Han dynasty) that means something similar to “over look, not pay attention,” then you'd have a stronger case. I did do a simple search on www.ctext.org and didn't find any. None of the sources I looked at showed 勿 as having a meaning.

Ex. 4: 漫 flood, all over
* water
* long, extended

Analysis:
Though not its original meaning, the meaning “long” for 曼 was probably in existence when the character 漫 was invented. I'm not sure that “long” is related to the notion of “flood.” “Width” and “depth” are at least as related and I would think “depth” is much more closely related than “long.” There seems to be a better case for “extended” as in “extended waters” = “flood.” You might be right in this case, but I'm not sure. None of the sources I looked at (and I looked at quite a few) showed 曼 as having a meaning.
 

rizen suha

状元
Thank you very much. Your analyses are extremely interesting and enlightening. They resolve all my doubts and questions. I am anxiously looking forward to using the future editions of the outlier dictionary. Judging by the level of insight that you provide here, the dictionary will be a true treasure tome of knowledge. Good work!
 

Abun

榜眼
The original meanings of 冊 cè and 篇 are very close. Both are writings on bamboo strips that are tied together. The original meaning of 扁, though it's related to writing, is still a bit different. It did not have the meaning “flat,” though it could mean “essay,” but not the kind that is written on paper. None of the sources I looked at showed it as having a meaning.
I’m guessing with “paper” you refer to writing material in general, seeing as paper would be invented until the first millennium AD, right? But in that case I’m not quite sure what kind of essay you are referring to. Do you mean 扁 was more the abstract concept of the essay, the rhetorics, the line of argumentation etc, rather than its physical written form?

Either way, it sounds like 扁 and 篇 are extremely close in meaning. In your opinion, are they etymologically related? And could it maybe even be the case that 扁 in its very earliest form did mean a text written on bamboo strips as its components suggest, and only later people added the bamboo radical to write this meaning while the original character got reappropriated to a derived, more abstract concept (similar to what happened with 文 and 紋)?
 

Ash

进士
>I’m guessing with “paper” you refer to writing material in general, seeing as paper would be invented until the first millennium AD, right?
I wasn't referring to writing material in general. I said, "not the kind that is written on paper." (i.e., because paper wasn't invented at that point in history).

>But in that case I’m not quite sure what kind of essay you are referring to. Do you mean 扁 was more the abstract concept of the essay, the rhetorics, the line of argumentation etc, rather than its physical written form?
扁 doesn't mean "essay," 篇 means "essay." 扁 means “large characters written on a sign or placard and posted somewhere.”
I would think that 篇 referred to shorter writings, while 冊 referred to longer writings.

>Either way, it sounds like 扁 and 篇 are extremely close in meaning. In your opinion, are they etymologically related?
To say something has an etymological relation, there usually has to be a strong meaning and strong sound connection. Sound-wise, 篇 and 扁 are close. Meaning-wise, they do have a relation: "essay" vs. "writings posted on a door (or whatever)." Both are types of writing. The spoken words that 篇 and 扁 represent are probably etymologically related, but I still don't think that 扁 is giving a meaning in 篇, because I don't see how "large characters posted on a door" is related to "essay."

>And could it maybe even be the case that 扁 in its very earliest form did mean a text written on bamboo strips as its components suggest, and only later people added the bamboo radical to write this meaning while the original character got reappropriated >to a derived, more abstract concept (similar to what happened with 文 and 紋)?
This type of thing does indeed happen, but probably not in this case. 冊 and 篇 have very similar meanings, like I said, but 扁 is composed of 戶 "door" and 冊 "writings on bamboo strips," which doesn't really suggest the meaning "essay."
 

rizen suha

状元
Historically, at the moment of creation of a character:
When choosing a sound component, could it be possible that one with a "close enough" meaning was preferred?
The more available syllables, the higher the possibility of a good fit.
Therefore sometimes the fit seems extraordinary clear, on other occassions a bit of imagination is required.
 

Abun

榜眼
扁 doesn't mean "essay," 篇 means "essay." 扁 means “large characters written on a sign or placard and posted somewhere.”
Ah, I misunderstood then, thanks for the clarification.
>Either way, it sounds like 扁 and 篇 are extremely close in meaning. In your opinion, are they etymologically related?
To say something has an etymological relation, there usually has to be a strong meaning and strong sound connection. Sound-wise, 篇 and 扁 are close. Meaning-wise, they do have a relation: "essay" vs. "writings posted on a door (or whatever)." Both are types of writing. The spoken words that 篇 and 扁 represent are probably etymologically related, but I still don't think that 扁 is giving a meaning in 篇, because I don't see how "large characters posted on a door" is related to "essay."

>And could it maybe even be the case that 扁 in its very earliest form did mean a text written on bamboo strips as its components suggest, and only later people added the bamboo radical to write this meaning while the original character got reappropriated >to a derived, more abstract concept (similar to what happened with 文 and 紋)?
This type of thing does indeed happen, but probably not in this case. 冊 and 篇 have very similar meanings, like I said, but 扁 is composed of 戶 "door" and 冊 "writings on bamboo strips," which doesn't really suggest the meaning "essay."
Yes, I should have worded my question more clearly; I was of course asking about an etymological relationship between the words which the characters 扁 and 篇 represented at the time of their creation, not the characters themselves. In any case, thank you very much for the interesting insight.
 

Ash

进士
Historically, at the moment of creation of a character:
When choosing a sound component, could it be possible that one with a "close enough" meaning was preferred?
The more available syllables, the higher the possibility of a good fit.
Therefore sometimes the fit seems extraordinary clear, on other occassions a bit of imagination is required.

We will probably never know for sure, but my impression, after having examined many of the phenomena that happen with Chinese character evolution, is that people back then didn't sit around and ask themselves questions like “Can I find a sound component that also has a meaning?” “Is the meaning of this sound component close enough to the character's meaning?” Rather, when a sound component also conveys a meaning, it's mainly because what we're now calling the sound component is the original version of the character and another semantic component was added later.
Ex. 莫 “not; do not” is the original character for 暮 “sunset.” 莫 was originally a 'picture' of the sun between grass (i.e., what you could only see at sunset – or sunrise). The bottom part got corrupted into 大 in the modern form. This form was taken over by a similar sounding word meaning “not; do not.” So, a new character was created to represent “sunset” by adding a 日 “sun” to the original character.

Ex. 現 xiàn "to appear" was originally written as 見. In ancient texts, 見 could represent either xiàn "to appear" or jiàn "to see.” Note how close these words are in sound and meaning. Eventually, a 玉字旁 was added to the original character so that xiàn "to appear" could have its own character.

It is possible that some characters were invented in the way that you describe, but I would think that those are characters that appeared late in history and are few in number. It's important to note that there are an infinite number of stories that you can assign to any character. The issue is, if you're just making up stories, you're going to end up hiding the real sound and meaning patterns that are there. It's exactly those patterns that help you learn more effectively. As such, our job is to find out the real story for each character (i.e., uncover the real sound and meaning patterns), insofar as that's possible given the evidence available. If I don't see strong evidence that a given sound component is actually giving a meaning, then I'm not going to say that it is. I'm always willing to change my opinion, but that needs to be based on evidence and not just mere possibility. Just because something is possible, doesn't make it probable.

Without having done any systematic study on the issue, I would think that the cases where “the fit seems extraordinary clear” are situations like the examples given above and the cases where “a bit of imagination is required” are just that -- imagination. : )
 

rizen suha

状元
Thanks for guiding us expertly through these labyrinths of history and contemporary rationalization. At every turn, a surprise awaits. In your research, in addition to going through throves of written sources, do you also get to consult "face to face" with the epitome of (adequately gray haired) chinese scholars?
 

rizen suha

状元
... my impression, after having examined many of the phenomena that happen with Chinese character evolution, is that people back then didn't sit around and ask themselves questions like “Can I find a sound component that also has a meaning?” “Is the meaning of this sound component close enough to the character's meaning?” ... It is possible that some characters were invented in the way that you describe, but I would think that those are characters that appeared late in history and are few in number.

This is quite surprising, I feel. One would have thought that the natural inclination of the scholars of their time would have been to optimize the semiology of the characters.
 

Ash

进士
Thanks for guiding us expertly through these labyrinths of history and contemporary rationalization. At every turn, a surprise awaits. In your research, in addition to going through throves of written sources, do you also get to consult "face to face" with the epitome of (adequately gray haired) chinese scholars?
I do to some degree get to consult "face to face" with Chinese scholars. I was introduced to paleography by 陳劍 (took two 2-week courses from him in 2007 and 2009) and learned it more thoroughly from 季旭昇. I took quite a few graduate courses from him and went to his weekly excavated texts study group for several years. I also took a class from 杜忠誥 (as did John). Lately, though, I haven't spent much time with them, because of the demands on my time due to Outlier. I've met another hand full of scholars at various paleography and sinology conferences.
 

Ash

进士
I guess that for most Chinese scholars (specially gray haired ones) the long tradition carries much more weight than the new evidence.
That's fairly true, though there are a number of gray haired scholars that are rather open to new things. I basically only make use of scholarship from people who I perceive to base their ideas on evidence. There are some out there that are basically just trying to defend the traditional way of thinking about things from new evidence.
 
Is it worth getting both traditional and simplified versions of the dictionary? How many unique entries does one have as compared to the other?
 

Shun

状元
I bought the Traditional one and automatically got the Simplified one for free, as well. This was on an iPhone. Both have about 1,850 entries.
 
Top