I’m trying to read a book about Go (the game) and there is one kanji that I can’t find a reference even if I draw it on Jisho or Google translate. I will describe it using Wanikani (pre-reform) radicals: there is a “nailbat” and next to it, at the top, there is a “triceratops”, below the “triceratops” there is the “ground”, below the ground there is a small “mouth”, and below the mouth, there is a “rice paddy”… Thanks for any help.
With your description and google translate in chinese mode, I found 擋. Is is the right one ?
4 Likes
yes thanks!
There are apps that let you find a kanji by choosing the component radicals. I like this one for iOS:
Here is what the radical picker looks like:
2 Likes
Or Jisho (just push the “radicals” button to the left of the search bar).
1 Like
I tried that and did not work for this kanji. When I look up this kanji directly on jisho, the only radical it lists is the hand and no components are listed.
It never works properly. 
Are you sure it’s the right kanji? I do not get any relevant results in Japanese on google.
It’s also not on kanjijiten nor kanjipedia…
Maybe it is a name kanji? Does it have furigana the first time it shows up?
1 Like
Yes, later I was able to find on IMIWA (Kanji dictionary of iOS). It means obstruct, impede, stop, resist, etc… which matches the context on the Go problem I was reading (white invades and black can not stop, etc…) On reading: トウ and kun reading さ (さける）. Jisho has the same information as IMIWA but the entry for parts is empty. Using IMIWA, I could get the kanji by typing the nailbat and the triceratops. 擋 - Jisho.org
The Go book in which it appears is all vertical writing with zero furigana.
1 Like
Well, imiwa and jisho.org both use the same dictionary files (Jmdict & friends); jisho.org is probably using an older/outdated version, I guess. If you look at the kradfile, which provides the base data (I don’t think they use anything else), you can see the following decomposition:
擋 : 冖 口 尚 扎 田
P.S.: For what it’s worth, I found it pretty easily on my denshi jisho using the handwriting input, whereas I could not get any of my dictionary apps to recognise it. 
Cool! I did not know about the RADKFILE/KRADFILE!
I used to think that WK’s radical names weren’t sticking, but still somehow managed to perfectly visualize the kanji as described. . .
3 Likes
Usual warning that JMDICT isn’t a (completely) trustworthy resource.
It also “knows” 妛, which is a certified fake kanji. (It makes you wonder how it even got a reading for that one)
If I cannot cross-reference something from JMDICT, I just ignore it.
What do you mean by “certified fake kanji”? It also has a Japanese section on wiktionary. Do you just mean it’s not used in Japanese?
1 Like
?
@seanblue I followed the same steps. 
1 Like
It’s one of the main ghost kanjis that exist in shift JIS. For this one, they were able to find that it was introduced by an error in a photocopy machine. So yeah, it’s fake.
I’m surprised you ask that, though, since there was a whole thread on the forum about those kanji and I think you interacted there.
4 Likes
Yes, but, is there a certificate?
3 Likes
Does the official report from the JIS group count as certificate?
1 Like