View Single Post
Old Mar 9, 2011, 05:28 PM
dancey's Avatar
dancey dancey is online now
Trusted Editor
Join Date: Dec 2007
Location: New Jersey
Posts: 1,266

This guide is intended for those with little to moderate understanding of the Japanese language. It is by no means comprehensive and may very well be inaccurate in areas. Please feel free to correct or improve portions of it.

An Illustrative Guide to Radical Lookup


Software Used in this Guide: jQuickTrans

Kanji, literally translated, means chinese characters and are an integral part of the Japanese writing system. You can read all about the history, variations and the differences between Chinese and Japanese kanji here.

Radicals are the components of kanji characters. Each individual kanji can be made up of one or more radicals. There are 7 basic radicals which vary based on the position in which the radical is located.
  • hen (left side)
  • tsukuri (right side)
  • kanmuri (top)
  • ashi (bottom)
  • tare (left+top)
  • nyo (left+bottom)
  • kamae (box)

There are 200+ different radicals, each with it's own definition. The definition of a radical can often times hint towards the definition of the kanji it builds. For more information on radicals, click here. For an index of radicals, click here.

You can rebuild any kanji using radical lookup and, in the context of this site and this guide, you can use it to lookup single characters, words, recreate entire sentences or pages or anything you're interested in understanding. That being said, for someone with a basic understanding of the Japanese language and unfamiliarity with different radicals, this can be a very time consuming process. It is most useful for looking up (or verifying) proper names for album credits when you cannot find a Japanese webpage with credit information either due to the age or popularity of the album.

Looking up Radicals with jQuickTrans

In jQuickTrans there are 3 main sections of the program. The topmost section is the input area, the bottom left section is the kanji info section and the bottom right section is the dictionary section.

For the purpose of this guide you'll want to focus on the lower left side of jQuickTrans. By default, the radical lookup pane is hidden, so you will need to enable it by clicking the "Radical Lookup" button next to the Print icon on the kanji info section or by clicking Options --> Kanji Info --> Radical Lookup.

The radical lookup section mirrors the radical index numbers noted in the wiki link above, which is ordered by stroke count. The numbers within the kanji info pane indicate the number of strokes required to draw that radical.

All you need to do to lookup kanji is click on a radical (or a combination of radicals) in the radical pane and the kanji info section will populate with kanji that have that radical or combination of radicals.

So let's start with an example. I'll use "Sakimoto Hitoshi" ("崎元 仁") as a simple example. You may want to increase the default 200 kanji results to something higher in order to pull more results. This is particularly useful if your kanji only has one or two radicals because your kanji info pane could be populated with a large number of kanji. It is also useful to sort by the column labeled "Frequency", the lower the number the more frequent that kanji appears in writing and common dialogue.

Starting with the left-most kanji you can immediately recognize 3 radicals: the "山" left-most radical, the "大" right-top kanji, and the "口" on the bottom right. I find it easiest to start with radicals that I recognize first (often the simplest) and work my way towards harder/more complex radicals as the list becomes more and more filtered. So the most simple radical I see in that is "口", so find the "口" in the radical list and click it. You now have a populated list of kanji in your info pane that contain the "口" radical.

Now let's move on to another radical that is easily recognizable, the "山" shaped radical. So find the "山" radical in the radical pane and click it. You can see the size of the list has been noticeably reduced by the size of the scroll bar. You can also see that at the top of the list, since you sort by frequency, you see the kanji that you're looking for ("崎")! You don't even need to enter the other radical(s).

Once you find the radical you're looking for, you can click and drag the kanji from the kanji info section to the input section to save it for later use or to build more of the word(s) you're trying to build.

So the same process applies to the second kanji. Press the "x" button to clear the radical lookup results and start by trying to identify some radicals in the kanji. In this case, you have a very simple kanji and these kanji are often the ones that are most difficult to find. You see there is a "一" radical and a "几" radical. If you click the "一" radical, you'll see that there are literally thousands of possibilities for this radical. So now you need to look for the "几" radical. If you click that, your list is filtered to one result, wait what? That's not the kanji you're looking for! That means you've picked the wrong radicals, so let's go back to our radical index and see if you can find a better match. Often times this means either taking what you thought was a complex radical and breaking it down into multiple simple radicals or taking multiple simple radicals and combining them into a more complex radical.

So you had two relatively simple radicals, clear the list and let's start from scratch! Let's try to break it down into even more simple radicals. You can't break the single line radical down any further so maybe you can break down the "几" radical. If you look in the radical index there is a "mans legs" radical (儿), which could work. Click that and let's see what happens. Looks like right off the bat you got a better hit than you did using the "几" The 5th most frequent radical (先) is really close to what we're looking. Let's try clicking the "一" radical again. This time your list is much shorter but again, you don't see anything in the list that resembles what you're looking for. Let's try removing the "一" radical and going back to the kanji to identify additional radicals. All we have left are the line at the top and the line above the "儿" radical, so maybe we can combine these two radicals into a "二" radical. If you click the "二" radical and look in the list, the 3rd most frequent kanji is exactly what we're looking for ("元"). Now you've rebuilt the last name and can use that for searching ("崎元").

The third kanji follows the same principle. You see 3 radicals in this one, two "一" (or maybe one "二") and "彳" We already know from the previous kanji that maybe it's a good idea to try and start with "二" radical, so filter on that first. There's still a lot of kanji in the info panel, though. If you scroll down the list a bit you will find the kanji we're looking for. You can also filter further by using "彳" and you see "仁" in the 4th most frequent position.

Now that we've rebuilt the radical you now have something you can type into search engines and hopefully find out how to pronounce it. Do not rely on the jQuickTrans dictionary for name translation. You can actually change the dictionary from "edict" to "Proper Names" and get names for the kanji in the input section but for many Japanese names there can be a LOT of variation in pronunciation. It's best to try and find websites that clarify the kanji with hiragana to get a correct romanization.


The more you lookup kanji using radical lookup the more familiar you will become with the individual radicals themselves and it will reduce the time you spent trial-and-erroring and scrolling through the kanji info list.

Some things to note as I've been doing this for a few years now:

- The "一" radical pretty unreliable. Sometimes I see it included in kanji that it shouldn't be included in at all (?) and it's just so simple it's often misleading as to the true representation of the radical.
- Some radicals have a vastly different individual radical representation than how they are represented in kanji. The "彳" radical is always used on the right but the straight radical representation is misleading (at least in the application) if you are doing this by character recognition alone.
- There are a handful of complex common radicals in the 6-9 stroke range that are misleading if you're not familiar with radicals. A good example is "自" and "首". You can spend a lot of time piecing together radicals and never finding a kanji if you're not using the right ones, so if all lookups keep coming up empty, try looking at more complex radical combinations.
- There are a handful of simple kanji you should familiarize yourself with so you don't have to do radical lookups for single radical kanji. Good examples are "一", "八", "十", "几", "中", and "口". These can be a pain to lookup because you end up having to scroll through hundreds of kanji looking for something simple.
- Often times if I'm having a hard time identifying radicals I will put in one simple radical in the kanji and scroll through the list and see if I can find anything that resembles the kanji I'm looking for and look at the radicals that it's composed of. This might clue you in to a radical you're not identifying properly.
- The jQuickTrans program it looks like isn't really being supported or maintained anymore (hence freeware... I think I paid 30$ for it 7-8 years ago). I've noticed some rather unfortunate bugs in the software using Windows 7 likely related to comparability and such. There is a clipboard link option which is helpful because it will automatically pull your clipboard into the input area, but sometimes it just stops working. There's also some error messages every once and a while. Just reload the app and you should be fine.

As an aside, you can use these instructions to rebuild kanji in any application and I believe this is the same way you lookup kanji in a dictionary, though I've never personally used or have seen one.

Hope this helps some users who are as unfamiliar with Japanese as I am.
Reply With Quote