Alphabets

This page contains information about the kana (Japanese phonetic alphabets) and kanji (the Japanese ideograms), as well as a little bit about pronunciation, typing, romanization, punctuation, and honorifics. Though it's largely up to you when you start studying kanji, everything else on this page you should know before learning any grammar. In particular, you should know all the kana by heart. The rest of the guide will assume you can read kana (and look up kanji) on your own.

If you think it might help you memorize the kana, download Kana flashcards.

Pronunciation
Hiragana
Typing in Japanese
Katakana
Obsolete Kana
Romanization
Punctuation
Honorifics
Kanji
Introduction
List of Radicals
Kanji Readings
Reading Changes in Compounds

Pronunciation

Compared to any European language, the pronunciation rules of Japanese are almost unbelievably simple, in that there is almost zero ambiguity. The basic difference between English and Romaji pronunciation is the vowels. If you happen to know Spanish, you’re in luck, because that language uses the exact same vowel pronunciation. If you don’t, here’s the short version:

English has at least a short and long sound for each, while Japanese only has one sound. The letter “a” can be “ah” or “ay” in English, but only “ah” in Japanese. “e” can be “eh” or “ee” in English, only “eh” in Japanese. “i” can be “ih” or “ee” or “ai” in English, only “ee” in Japanese. “o” can be “ah” or “uh” or “oo” in English, only “oo” in Japanese. Finally, “u” can be “uh” or “uu” in English, only “uu” in Japanese.

Here's another attempt at explaining it:

And another:

Hopefully that makes it pretty clear. The only time this gets any harder is with the vowel blends, but even those are pretty easy to sound out. Consider “ai” for instance. Try sounding it out: “ah-ee,” and then say it faster and faster. It should start to sound just like the English pronoun “I.” All other blends are just as easy. If you can't figure out what two vowels would sound like together, then they probably don't make a blend and you should pronounce them separately.

The only other things that confuse Westerners are letters like “tsu,” but that's also pronounced exactly as it sounds. Try to convince yourself that the “t” in “tsunami” isn't actually silent and it shouldn't take too long. “dzu” and “dji” are weirder but also not hard once you see them get used a few times (especially since nowadays they're usually pronounced exactly the same as “zu” and “ji”).

Hiragana

The pairs of short lines above four of the last five rows are called 濁り nigori. The circles above the last row are called 丸 maru. Every letter with a nigori or maru is a variant on one that has neither, so it is suggested you memorize these letters in doubles and triples like かが and はばぱ whenever possible. Not counting variants, there are 55 hiragana.

A glide is a special combination of two kana whose sounds can blend together. All glides require the second letter to be written smaller than usual; if it isn't small pronounce them as two separate letters.
For example, 「きゃ」 is “kya” but 「きや」 is “kiya”

Also, a っ or “little つ” always means that the following letter has a double consonant.

For example: きっぷ kippu,　けっこう kekkou,　やった yatta

Note that the glides にゃ, にゅ and にょ can be easily confused with the non-glides んや, んゆ and んよ when romanized. For this reason the latter trio should be romanized with an apostrophe after the "n" to indicate the syllable break. To clarify:

にゃ nya ≠ んや
にゅ nyu ≠ んゆ n'yu
にょ nyo ≠ んよ n'yo

For example: けんいち = Ken'ichi, not Keni-chi or Ke-nichi

Typing in Japanese

If you can’t type in Japanese yet, add Japanese to your keyboard languages now and get used to the IME (Input Method Editor) right away. If you don't know how, explore the Language settings in your control panel, or use a search engine to find a thorough tutorial for your OS.

Once you've switched to Japanese input, first you simply type the romaji for each kana (i.e. type “ka” and か appears, type “mya” and みゃ appears). As you type, notice there is a dotted line underneath the kana. This means they haven't actually been entered yet because you may want to convert them to kanji. Press the space bar to make the IME guess what kanji you want. It will then split the string into several segments, convert them to kanji, and underline the farthest left segment. Press the space bar again to see a menu with other conversion options for this segment. Use the arrow keys to scroll through these menus or change which segment you want to see the menu for. Pressing enter at any time during this process (even before pressing space the first time) will remove all underlining and actually enter the string in whatever form it currently appears.

The details of conversion you can and should work out yourself by experimenting, or by googling for more thorough tutorials. However, there are some important exceptions to the “just type the romaji” rule, as follows:

To type a small hiragana or katakana outside a glide, either use the conversion menus (the long way), or type "l" or "x" in front of it. For example: typing "ya" makes や while typing "lya" or "xya" makes ゃ. The same trick allows you to type やよゆあえいおう as ゃょゅぁぇぃぉぅ for assorted aesthetic reasons, though it won't work on any of the other hiragana.

The following letters/blends are typed differently from how they are pronounced. These are mostly the result of non-Hepburn romanization systems, as explained below.

"tu" and “tsu” both make 「つ tsu」

"du" makes 「づ dzu」

"di" makes 「ぢ dji」

"dya" makes 「ぢゃ dja」

"dyu" makes 「ぢゅ dju」

"dyo" makes 「ぢょ djo」

"n’" makes 「ん n」

"nn" makes 「ん n」

Typing 「ん n」 can be particularly annoying because typing any consonant or punctuation mark after it will cause it to convert to ん even if you did not type "n" twice. This makes it easy to get in the habit of typing "n" once, which usually results in problems like typing にゃ when you meant んや, or having words end with ｎ for no apparent reason. The ideal habit is to always type “nn” for ん just to be safe.

Katakana

A loanword, if you don't already know, is a word in one language that speakers of another language borrow and start using themselves, to the point where it becomes part of their language as well, often with a different meaning from the original.

When using katakana to write a loanword which — in the original language — ends in a consonant, the letters "ku," "su," "ru," "to," "fu," and their variants (gu, zu, do, bu, pu) will often have their vowels dropped without any extra indication.

All the hiragana glides work the same in katakana, with the following additions:

キャ kya now also means ca (like in "cat")

ファ fa フェ fe フィ fi フォ fo

ウィ wi ウェ we ディ di

ヴァ va ヴェ ve ヴィ vi ヴォ vo (the kana ヴ for "v" is a very recent invention, so a "b" letter will just as often be used to imitate the "v" sound)

The chouon (dash) is a shortening of the term 長音符 (ちょうおんぷ) chouonpu, literally meaning "long sound mark." It is used like this:

アー er/ar/or (also applies to all other letters with an "a" sound)

オー o (specifies a long "o," also applies to all other letters with an "o" sound)

シー shii (specifies a long "I," also applies to all other letters with an "i" sound)

These symbols can be used to stand for sounds that Japanese doesn't use. When speaking Japanese, stick to the pronunciation these letters would normally represent even if they are being used to imitate English words.

シ shi (“see”; like in "season")

ス su (hard “th”; as in "health")

ザ za (soft “th”; as in "the")

ン n (“m”; only if followed by another consonant)

ツ tsu (“tu”; like in "two")

These two symbols are also very common in loanwords and foreign names:

・ (represents where a space would have been placed in English; typed with the / key)

＝ (used when a given name is written before a surname)

Since loanwords are almost always written in katakana, here are some random examples to help clarify the rules from part 2. Many of the rules listed above affect pronunciation but not romanization, so the romaji here may not reflect them.

システム shisutemu (system)

レベル reberu (level)

ビデオ bideo (video)

パウワーアップ pauwaa appu (power up)

ザファースト za faasuto (the first)

マイスイートハート mai suiitohaato (my sweetheart)

ホワイトクリスマス howaito kurisumasu (White Christmas)

パソコン pasokon (PC) = パソナルコンピュター pasonaru konpyutaa (personal computer)

Obsolete Kana

Since these aren't used anymore, you certainly don't need to know them, much less waste time memorizing them, but it's probably a good idea to know they exist.

Romanization

Romanization means taking a language not written in the Roman alphabet, and somehow writing it in the Roman alphabet.

I use something close to Hepburn style for all of this guide, because that's the system designed to actually look like the correct sounds to native English speakers. That's also the style you saw above in the kana charts. The other popular systems, Kunrei-shiki and Nihon-shiki, were made by and for Japanese speakers.

Things distinctive of non-Hepburn styles include romanizing つづしふじ as tu, du, si, hu, zi, and writing all glides with a “y” in the middle (e.g. じゃ as “zya” instead of“ja”) Although sometimes ふ should in fact be read as “hu”, most of those spellings are simply misleading to any Westerner trying to learn the language. You need to be aware of them simply to avoid being confused when you do see them, because they do get used a lot (especially in Japan).

Even within Hepburn, there are tons of variations and controversy. Most notably, my usage of dzu/dji for づぢ is technically obsolete, but I'm convinced it's useful, if only because using the same romanization for two distinct kana (zu for both ずづ or ji for both じぢ) is just asking for confusion.

Punctuation

In Japanese, punctuation (including spaces) is entirely optional, so usage is a bit more inconsistent than in English, but a lot of it is basically the same. Just don't rely on it to tell you when clauses and sentences end as much as you would in English.

The marks to know are the period (。), the comma (、), the square brackets (「」) which act as quotation marks, and the hollow square brackets (『』) whose exact meaning are entirely up to the author. Also notice that the characters themselves have a lot of blank white space, so even if it looks like Japanese text has spaces after its commas or periods, it really doesn't. The reason for this filler is that every single character in Japanese is meant to fit within a square.

Honorifics

In English, our honorifics include the prefixes Mr., Ms., Miss, Sir, Ma'am and occupational titles like President or Supervisor. While occupational titles can always be looked up and often translated literally, the most common of the general purpose Japanese ones need to be explained. Also, all Japanese honorifics are suffixes.

-[blank]: unlike English, simply not using an honorific (when addressing someone directly, doesn't matter in the third person) implies a very close relationship

さん -san: the default honorific for just about anyone, regardless of age or occupation

君 -kun: a more friendly honorific, usually used for males of the same age or lower

ちゃん -chan: a friendly and cute honorific, usually used for young females

様 -sama: a respectful honorific used for those of a higher social standing

殿 -dono: even more respectful than -sama, less likely to be used solely out of obligation

These last three have English equivalents, but are so incredibly common that they often can't be translated literally (or at all) without sounding far stiffer than they should.

先生 -sensei: usually “teacher” or “doctor,” also any professional the speaker respects

先輩 -sempai: “upperclassman,” someone with more experience or in a higher grade level

後輩 -kouhai: “underclassman,” someone with less experience or in a lower grade level

At this point you've read everything you need to know before moving on to Grammar Part 1: Particles. I would recommend skimming the kanji information below so that you're not completely unaware of it, but you don't have to seriously study the kanji for quite some time (since you can get source material with furigana).

Kanji

Introduction

漢字 kanji are ideograms. In other words, they represent ideas. Not sounds, not words, but ideas. That's the number one thing I want you to get out of studying them. Often, they are also combinations of other ideograms, which allows you to pull them apart and come up with some kind of rational explanation for these arbitrary shapes. The simplest elements a kanji can be broken into what I call primitives. Traditionally, the Chinese organize their hanzi and the Japanese their kanji using a standardized list of 214 primitives (and their common variants) called radicals. If you have a dictionary that lets you look up kanji using their radicals (and you should before you study the kanji) then you can use it to look up each radical's meaning on your own.

A quick warning: don't ever rely on lists of kanji to tell you what you should learn. That's just as silly and counterproductive as thinking that there can be a list of a couple thousand English words you need to know to use the language, and nothing else. Lists can certaintly be good starting points, but never treat them as anything more than that. The majority of your kanji knowledge will come from reading actual source material, especially when it comes to things like knowing which reading or meaning to use in which compounds.

As implied earlier, if you can recognize what all of the radicals and small kanji in a really messy kanji mean, sometimes you can guess what the entire kanji means just from that. Most of the time, you can't. The reason studying radicals is useful is that once you know them, you look up a kanji, then come up with a way of associating the meanings of the radicals with the meaning of the kanji as a whole, often in the form of some amusing nonsense sentence or pithy statement. Once you start doing that, remembering what complex kanji mean and look like gets far easier. A few quick examples:

魔（ま）demon/magic

Elements: 53广cave, 林（はやし）forest, 194鬼（おに）demon

Demons and magical creatures live in a cave hidden inside a forest that nobody dares to enter.

神（かみ）god

Elements: 113c礻altar, 102d申 speak

Priests speak to god at the altar.

騎（き）horse riding

Elements: 馬（うま）horse, 奇（くしき）strange

A wild horse would find horse riding to be quite strange

If that looked utterly arbitrary, it was. There is no general method of determining a kanji's meaning from its elements, so the idea here is simply to recognize kanji inside other kanji, and some really common radicals with meanings everyone agrees on, so that you can invent these mnemonics and save yourself a lot of time and effort. Of course, for this to work, you have to memorize several dozen simple kanji first. Once you've done that, take a look at the radicals list below and start looking for them the next time you try to read something in Japanese.

Finally, a brief touch on stroke counts/order. If you want to see the strokes for any kanji, visit http://kakijun.main.jp/main/kensaku.htmland type it in. You don't need to know anything about strokes to simply read kanji, but since radical-based dictionaries rely on stroke count to organize entries, and you have to know strokes in order to write any kanji by hand well, it's still good to know.

List of Radicals

Format is: radical number, stroke count in (), the radical itself (may be shown indirectly inside {}s), and meaning. If the radical can only appear on a certain part of a kanji, it's mentioned inside []s.

Most of these meanings are widely agreed upon, and all of them should be useful, but feel free to ignore or disagree with some of them since these are in no way hard rules. And of course, you do not have to memorize all of these anytime soon.

Kanji Readings

Now the important part. Just about every single kanji has at least two readings, or ways it can be pronounced. Most of these readings are either 音読み on'yomi or 訓読み kun'yomi, and it's safe to assume every kanji you will ever encounter has at least one of each. While there is no strict way of distinguishing the two categories, a lot of useful rules of thumb can be stated.

As an example, let's consider the very typical kanji 走. Its readings are:

1) the kun'yomi はしる, a verb meaning “to run” with okurigana る

2) the on'yomi ソウ meaning “run,” used in compounds like 疾走 sprint and 走者 runner

And in contrast, a potentially annoying one, 間:

1) the kun'yomi あいだ, a noun meaning “space/interval”

2) the kun'yomi ま, a noun meaning “space/room” or “time/pause,” used in many idiomatic phrases

3) the on'yomi カン meaning “period of time,” “space” or “between,” used in numerous compounds

4) the on'yomi ケン meaning a figurative “space,” used in the compound 世間 society

You may be wondering in cases like 間 how to tell which kun'yomi or which on'yomi to use. The short answer is there is no way other than memorizing all the words they're used in. The long answer is that once you've seen the kanji used for a while, you'll get a feel for when one reading or the other gets used more often, or know that this reading is only used in a handful of compounds or this reading is never used on its own, etc. But remember this is strictly on a case-by-case basis.

Note that both of those example kanji are a bit lacking in usage examples. It's mostly up to you to study and pay enough attention to the kanji that you can eventually remember most of their readings, what they mean and what words they're used in. Also, if you want to see how messy it can get, try looking up 生 or 下.

You may recall I said “most” (not "all") readings are either kun'yomi or on'yomi. The third category I was implying is 名乗り nanori, or name readings. Just as the term implies, these are readings which only get used in people's names. Many of these are so strange and irregular that entire dictionaries exist for nothing but nanori. However, since they're never used outside of names, you can safely ignore them for now and focus on learning real words instead.

Reading Changes in Compounds

When making compounds, some kanji can change their reading a bit. There are three (arguably just two) ways this can happen. To simplify the explanations below, pretend we're talking only about compounds of two kanji or two words.

1) If the last syllable of the first kanji is つ or less often く, it may change toっ

絶（ぜつ）beyond words ＋ 対（たい）versus ＝ 絶対（ぜったい）surely, absolutely

結（けつ）bind together ＋ 果（か）fruit, reward ＝ 結果（けっか）consequence, result, effect

抹（まつ）erase, rub away ＋ 殺（さつ）kill ＝ 抹殺（まっさつ）obliteration, denial

錯（さく）mix, disorder ＋ 覚（かく）remember ＝ 錯覚（さっかく）hallucination, optical illusion

2) If the first syllable of the second kanji is a kana which could have nigori added to it, then those nigori might be added. Same for marus.

死（しに）death＋神（かみ）god＝死神（しにがみ）shinigami, death god

番（ばん）a round, a shift, a number＋組（くみ）take part in＝番組（ばんぐみ）TV program

出（で）go out＋口（くち）mouth, passageway＝出口（でぐち）exit

3) Arguably not an actual change, but when the last syllable of the first kanji is ん and the first consonant sound of the second kanji is “m,” “b” or “p,” the ん might be pronounced more like an “m” than an “n.”

簡（かん）brevity＋明（めい）clarity＝簡明（かんめいkammei）conciseness

Next is Grammar Part 1: Particles.