An old seadog, a speaker of Japanese, returns to the port of RTK after many years absence adrift. He is tested by the KEYWORD--CHANT. Yes, he knows that-- a pyramid of three suns. "No," chants the taskmaster. As reference to RTK frame 21 reveals, the kanji is of a MOUTH next to two SUNS.

Question for RTKers: the multiple meanings of the graphic for DAY (RTK 12)--is that not an important factor in memory recall? That graphic has DAY, SUN, TONGUE WAGGER (speaker, etc.) and maybe others in competition. Ought primitive elements or the like be used consistently?

One person in this forum whose name escapes me suggested that variable meanings be individually anchored or distinguised by the physical location of the related element in the kanji (top, bottom, left, etc.). That's a good idea. Can we pool our resources together to iron out the RTK elemental meanings? Sort of like molding RTK, if it were a tree, into becoming a bonsai, if you get my drift, ha, ha.

I'm afraid this effort is doomed to failure. We'd like the Japanese language to be consistent in a way that would permit that sort of analysis, but the fact is that there are too many exceptions. In this example, the different meanings come from previously different kanji that have been simplified into indistinguishably similar kanji. So, the various meanings cannot be anchored or distinguished by any factor that I can think of. Not physical location or form anyway.

I'm starting to wonder if it might not be valuable to have a different name for the same primative when it is written differently. For example, the primative king, or heavy, they both take on a slanted final stroke when appearing to the left of something else. Well, sometimes they do. If it were consistent, then that rule would be enough. You actually have to remember when it happens and when it doesn't. For example, does it happen in "indications" or "delicate"? Or not?

CharleyGarrett Wrote:For example, the primative king, or heavy, they both take on a slanted final stroke when appearing to the left of something else. Well, sometimes they do. If it were consistent, then that rule would be enough. You actually have to remember when it happens and when it doesn't. For example, does it happen in "indications" or "delicate"? Or not?

To me this seems like one of the rules that is actually the most consistent in the kanji. Whenever something appears to the left and has a final horizontal stroke (soil, ivy, king, cow, fingers, metal, stop, etc.), final rightward hook (human legs, wind, spoon (as in "compare" 比)), or final drop (water drops, ice, abbreviated turtle), it is slanted upward. This makes sense when you think about the writing and stroke order. You are slanting up so that you can more easily write the next stroke which will be at the top.

Both "indications" 徴 and "delicate" 微 have the slanting up feature. There are some primitives that themselves have a hook that is added sometimes (parade in #387 "surpass" 越) or dropped (scarf in #402 "distant" 遠), but in combinations are there really any exceptions?

[ファブリス]

kanji Wrote:Question for RTKers: the multiple meanings of the graphic for DAY (RTK 12)--is that not an important factor in memory recall? That graphic has DAY, SUN, TONGUE WAGGER (speaker, etc.) and maybe others in competition. Ought primitive elements or the like be used consistently?

There's only two or at most three meanings for the primitives in RTK. "Day" and "sun" are really close. "Day" is the meaning of the kanji. As Heisig suggests, if it can help you build a more memorable story, you can use the kanji meaning instead of the primitive meaning where a kanji appears as a component of another kanji. A "tongue wagging in the mouth" can be much more helpful in many kanji than day, because "day" or "sun" is quite generic and could become vague if used in every kanji it appears in.

Does alternate primitive meanings complicate recall ? Not in my experience. Differentiating between the meaning of the kanji on its own, and its meaning as a component dit not bother me much.

Not everything is perfect. There are tradeoffs to be made. Two meanings for one primitive is a tradeoff to help in creating stories.

CharleyGarrett Wrote:For example, the primative king, or heavy, they both take on a slanted final stroke when appearing to the left of something else.

Part of it is logical, i.e. most of the time, when on the left, a bottom horizontal stroke will be drawn with the brush moving from bottom left to top right, going in the direction of the next components to draw, in [kana]gyousho[/kana] it may not even leave the paper.

Part of it comes from a feeling for composition I guess. Even if you're not an artist I think you can develop a feel for it, and there are also patterns that can be found in the kanji where certain elements get drawn the same way when they are paired with certain other elements.

In this case, for indications you can see that the last stroke of the *king* would collide with the third stroke of the *taskmaster*, either the taskmaster's leg (the third stroke) would have to be shortened, or the last stroke of *king* would need to go up, making the king look cramped and too small, furthermore the horizontal bottom stroke of king would not look nice with the curved "leg" from the taskmaster. In this way, the last stroke of king follows the taskmaster's third stroke. You can also see this pattern in several other kanji, where when possible the third stroke of the taskmaster extends to the left, and sometimes a little below other primitives.

Imagine that each radical was clearly spaced from one another, you could see vertical or horizontal spacers between them. This would "break" the kanji, making it look more like a grid than a single entity; especially if you stick several kanji side by side.

ファブリス Wrote:Not everything is perfect. There are tradeoffs to be made. Two meanings for one primitive is a tradeoff to help in creating stories.

Being able to make primitives more versatile by giving them alternative meanings is essential to me in making meaningful and memorable stories. I have absolutely no problem remembering that "day", "sun" and "wagging tongue" all equate to 日, none whatsoever. I notice Heisig offers these variations sparingly and usually for primitives that have to play wide roles. They appear in such a wide variety of kanji that to have to use, say, "sun" in the story for every one would tax my imagination. I occasionally add my own auxiliary meaning to primitives, whilst keeping the originals. For example, I find "computer" limiting for 里 so I use my own meaning of "home village" (the 訓読み　さと　suggests this meaning) in a number of kanji that use this as a primitive, but I still use "computer" in others. I wouldn't be interested in any scheme that sought to limit primitives to one keyword.

On the other hand I would be interested in hearing new meanings for primitives according to their position in the kanji. Heisig provides different meanings where the basic primitive has different forms according to position, 心 and 衣 being good examples, but primitives like 日 and 木 get into lots of positions without much change in form.

I gave a bad example for what I was thinking of. Actually, a better example is scarf. It goes all the way from "normal--just without the top hat of garment" all the way over to "rag"--which is actually enough different to HAVE another name. But sometimes the vertical stroke is more in the middle of the primative, and then it might not have a hook at the bottom of that vertical stroke. Do you see what I meant to say?

OK, I see what you mean. I checked and the kanji that have the other form of scarf in RTK I are "distant" - 遠, "park" - 園, "ring" - 環, and "send back" - 還.

To me it looks more like the last four strokes of "cloak" in this form. Maybe some word halfway between a "scarf" and a "cloak"? A "shawl"?

The other thing that might help to remember the form is that the first two are the "lidded crock with scarf" combo, and the latter two are "eye on its side, ceiling, mouth, scarf" combo. At least in RTK I, these combinations never have the normal version of "scarf". There's no rule that you can't make up your own primitives for these combinations Now with only 2 kanji for each primitive it might not seem worth it, but at least "ring" and "send back" are pretty complex otherwise. Anyway, I'm getting off-topic, but I think there could be value in giving slightly different drawings of a primitive a slightly different primitve meaning to help keep them distinct, but only when the change in the drawing isn't predictable.

Shvegait Wrote:OK, I see what you mean. I checked and the kanji that have the other form of scarf in RTK I are "distant" - 遠, "park" - 園, "ring" - 環, and "send back" - 還.

Correct me if I'm wrong, but "Ring" 環 is an exception to the rule that the hook on "scarf" is only dropped when the primitive is enclosed underneath (I don't think there is any other kanji where this rule is broken, in RTK1 or otherwise).

I think given that the "ring" combination appears only in "ring" and "send back" it shouldn't be impossible to remember the exception, you could even imagine it has an invisible "ring" enclosure round it which causes it to lose its hook.

Thanks! This discussion has really helped me!

Whoops, I forgot "monkey", which has "lidded crock with scarf" and nothing underneath. So scratch that Pangolin's explanation is good, but for me, since you draw the line underneath the scarf after you draw the scarf (and for "distant" and "send back" it's several strokes later), I find that it's easy to forget, and since the rule only comes up 3 times, and an exception once, I'm going to stick with "shawl".