When first encountering Japanese, it’s easy to be confused or overwhelmed by all the different characters and character types.
In this article, I’ll explain the different character sets to give you a clear understanding of what each type of character is for, and when each one is typically used.
Japanese has three main sets of characters:
- Hiragana – a phonetic set of characters unique to Japanese
- Katakana – another phonetic set of characters unique to Japanese, but used primarily for “loanwords”, or words borrowed from other languages
- Kanji – Chinese “picture” characters adapted to Japanese
In addition to these, there is also…
- Romaji – latin script (ie. English letters) used to represent Japanese phonetically
Ultimately all of these do the same thing:
Characters are a way to represent words in writing.
This seems like an obvious statement, but it’s helpful for understanding the Japanese writing system.
This is because most words in Japanese can be written in more than one way. Let’s take a closer look.
Hiragana and Katakana
Collectively called “kana”, hiragana and katakana are two separate, phonetic sets of characters.
Being “phonetic” means that each character represents a particular sound, and each character is always pronounced the same way (with a few exceptions). See my article on pronunciation for a more detailed explanation of all the sounds and sound combinations in Japanese.
There is both a hiragana and a katakana character for each and every unique sound in Japanese.
For example, the “a” sound is written like this:
“a” sound character
Here’s the full list for both:
Since both characters sets have a character for each and every unique sound, between hiragana and katakana, it is possible to write every single word in Japanese.
However, they are not usually interchangeable. The choice of which to use depends on which type of word is being written.
In essence, there are two types of words in Japanese:
- “Native” Japanese words – most words fall into this category
- Loanwords – that is, words borrowed from other languages (most of which are from English)
Native Japanese words can all be represented in writing using hiragana.
Loanwords can all be represented in writing using katakana.
For example, the word “karate” – a native Japanese word – is written in hiragana using the characters “ka”, “ra” and “te”:
On the other hand, “kamera”, the Japanese word for “camera”, is written using katakana since it has clearly been borrowed from English:
So it’s really that simple:
- If a word has been borrowed from another language, it is typically written in katakana.
- Otherwise, it can be written in hiragana.
There are two main exceptions to this:
- Katakana is used for some animals and birds, which often have quite complex kanji but are instead written in katakana, even though they are “native” Japanese words. “Frog”, for example, is “kaeru”, which is usually written in katakana (カエル) even though it is a native Japanese word that can be written in kanji (蛙).
- Katakana is sometimes used for native words in order to add emphasis, particularly in advertising. It’s kind of like a Japanese version of writing in ALL CAPS.
Kanji are Chinese characters adapted and applied to Japanese. They are generally only used for native Japanese words.
Remember how we said all native Japanese words can be written in hiragana? Well…
The majority of native Japanese words can also be written using kanji.
This brings us back to our obvious statement from earlier:
Characters are a way to represent words in writing.
Kanji simply gives us another way, in addition to hiragana, of writing most native Japanese words. For example, we saw earlier that “karate” can be written in hiragana, like so:
Well, it can also be written using kanji:
The way this is normally discussed is that every kanji has a “reading” – a way of pronouncing it – which can be written in hiragana.
So, as an example, the reading of “空手” is “からて”, or “karate”.
Kanji readings and “furigana”
Sometimes, kanji words are written with “furigana”, which is the hiragana reading on top of the kanji. Here is the word “karate” written in kanji with furigana:
Furigana is used, for example, when a kanji character or word is particularly rare, or has an irregular reading. Books and other material aimed at Japanese children and, to a lesser extent, teenagers, will also usually include a lot of furigana to ensure they can read the material.
In most cases, each sound that makes up a word is attributed to one of the kanji characters. For “karate”, the reading of the first kanji (meaning “empty”) is “kara”, while the reading of the second kanji (meaning “hand”) is “te”. These combine, of course, to form the word, “karate”.
There are also a few words with irregular readings where the reading applies to the word as a whole, so each sound can’t really be allocated to a particular character. An example of this is the word for “today”:
This is made of the characters meaning “now” and “day”, but the reading, “kyō”, can’t be divided and attributed to each of the kanji. It simply needs to be viewed as a whole, like this:
FYI, the Kana + Kanji Edition of my book, 80/20 Japanese, includes furigana for all words written in kanji throughout the book to ensure you can see the “natural” Japanese, but are also able to read it even if you don’t recognize the kanji.
When to use Hiragana, and when to use Kanji
Generally, most words that can be written in kanji usually are written in kanji. That means a word like “karate” will virtually always be written in kanji.
Overall, there are effectively three types of native Japanese words when it comes to their written representation:
- Whole word usually written in kanji
- Part of the word usually written in kanji, part written in hiragana
- Whole word always written in hiragana
Let’s look closer at each of these.
Whole word usually written in kanji
This type is simple as it is exactly as described – that is, the whole word is usually written entirely in kanji.
“Karate” is an example of this type. Most nouns also fit into this category since they only have one form (ie. they don’t conjugate like verbs), so their pronunciation is always the same.
Part of the word usually written in kanji, part written in hiragana
Many words, particularly verbs and adjectives, are typically written as a combination of kanji and hiragana. This is because only part of the word can be written in kanji.
For example, the verb meaning “go” is “ikimasu” in the polite form. In hiragana only, this would be:
However, this will normally be written like so:
The kanji, 行, only accounts for the part that is read as the “i” sound at the beginning. The rest of the word can’t be written in kanji, so it is always written in hiragana.
For grammatical reasons, all verbs that have kanji are like this – that is, they have part of the word that must always be written in hiragana. Many adjectives and adverbs are similar. Let’s see:
The adjective “new”:
The adverb “newly”:
There are also a few words that have hiragana sitting between two kanji. There are not nearly as many of these as the others, but they do exist, like this one:
This combines the stem of the verb for “eat” (tabemasu) with the word meaning “thing”. The “be” sound is actually part of the stem of that verb, and is always written in hiragana there too.
Whole word always written in hiragana
Some words simply don’t have a way to write them in kanji, so these are always written in hiragana.
For example, the verb for “do”, which in the polite form is “shimasu”, doesn’t have kanji, so it is always written in hiragana:
do (polite form)
This is true in its other forms too, such as the plain/dictionary form:
do (informal/plain form)
Words like this just don’t have kanji, so they are written in hiragana.
In addition, there are also some words that can be written in kanji but usually aren’t.
This is typically either because the kanji is too rare or difficult, or in a few cases, because it has actually been decided by the Japanese Agency of Cultural Affairs that certain words should always be written in hiragana.
A good example is the word “dekimasu”, which is the polite form of the verb meaning “can do”. This (as well as the same verb in its other forms) is virtually always written in hiragana, like so:
However, it does have kanji:
Individually, these kanji are both very common. The first is from the verb “demasu”, meaning “to leave, go out, or appear in”, while the second is from the verb “kimasu”, meaning “come”.
Despite this, the word “dekimasu” is basically always written in hiragana. The reason is because it has actually been officially designated to be written in hiragana.
There are literally thousands of unique kanji characters, but there is an official list of “joyō kanji” – kanji for common use – which outlines which characters should be taught in schools and used in official government documents. It lists 2136 different characters.
Here’s a link to the official searchable list (in Japanese, of course):
A lesser-known fact is that there is an annex to the list that essentially explains the general rules for kanji usage. Among other things, it also lists some exceptions – words that should be written in hiragana, even though the kanji that make them up are included in the list.
“Dekiru” – the plain form of “dekimasu” – appears as a specific example of a word that should, in general, always be written in hiragana. Here’s a snippet with the relevant part:
There’s certainly no need to memorize all of the rules and exceptions, but it does explain why you might occasionally see a word that you think should be written in kanji but isn’t. It’s not entirely random.
Summary of Hiragana, Katakana and Kanji
So, this all might sound confusing, but it’s actually quite straightforward.
- Loanwords (words borrowed from other languages) are written in katakana.
- Native Japanese words can all be written in hiragana.
- Of the native words in Japanese, some have kanji, and some don’t.
- When a word can be written in kanji, it usually is, whether that’s the whole word or only part of it.
- Some native words simply can’t be written in kanji, so they are always written in hiragana.
- There are also a few exceptions where hiragana is just preferred, even though kanji exists.
For the sake of clearing up any confusion, let’s finish things off by talking about Romaji.
Put simply, Romaji is Japanese written phonetically in Latin script (English letters). Its main use is to help people who can’t read Japanese.
It does have a few other practical uses, however. For example:
- It is sometimes used for emphasis in things like advertising and other signage
- It is used for some acronyms, particularly ones imported from English like “DVD” and “ATM”
- It allows Japanese to be typed using an English keyboard
Otherwise, Romaji mostly exists to help people who can’t yet read “real” Japanese.
Does that mean you should avoid using it?
Romaji is best thought of as a tool. It is extremely helpful for bridging the gap between zero understanding of Japanese and full reading ability.
It’s obviously great for tourists who visit Japan, as most signs with place names (train stations, highways, etc.) have romaji next to the native Japanese.
It’s also great, for example, if you want to learn to speak Japanese quickly without worrying about learning to read and write just yet. There’s no shame in this.
When you start learning Japanese, there is a lot of new information to learn – pronunciation, vocabulary, an unfamiliar grammar system, three new sets of characters… Romaji can make it all feel a bit less overwhelming, and allow you to get some early wins on your journey to speaking Japanese.
What should I learn, and when?
Now that you hopefully have a clear understanding of the different types of characters that make up the Japanese writing system, you’re probably wondering when the best time to learn them is.