The Jōyō Kanji List (常用漢字): The 2,136-Character Set Explained
The jōyō kanji list is the 2,136-character set that the Japanese government designates as a guide to kanji use in modern written Japanese.1 For learners, it is the closest thing Japan has to a national answer to this question: "Which kanji should an educated adult be able to read?"2
Overview
The list is published by the Cabinet of Japan as a 内閣告示 (Cabinet Notice), not a statute. It applies to general social use and four institutional domains: laws, official documents, newspapers, magazines, and broadcasting.3
What "jōyō" means in plain terms
The literal meaning of 常用 「じょうよう」 is "regular use." The 常用漢字表 (jōyō kanji table) defines itself as 「一般の社会生活において，現代の国語を書き表す場合の漢字使用の目安」 (a guide to kanji use when writing modern Japanese in general social life).3 For adults, it is advisory. For schools and government drafting, it is prescriptive. Together, these uses set the contract between writer and reader.4
The 常用 framing replaced the 当用 「とうよう」 ("for the time being") label of the 1946 predecessor. This signaled a shift from a permitted set to a regularly used set.25
The table explicitly excludes specialized fields and individual writing: 「科学，技術，芸術その他の各種専門分野や個々人の表記」.3 Proper nouns are also outside its scope. Prefecture-name kanji are the standing exception.3
The 2,136 number, fixed since 2010
The total of 2,136 字種 (character types) was set by 平成22年内閣告示第2号 (Cabinet Notice No. 2), issued on 30 November 2010.16 The notice added 196 characters to the 1981 list of 1,945. It removed five: 勺, 銑, 脹, 錘, 匁. The net change was +191.27
The same notice fixes 4,388 approved readings for those 2,136 characters: 2,352 on'yomi and 2,036 kun'yomi, plus 116 熟字訓 (compound-word kun readings).32 The 1981 内閣告示第1号 was abolished on the same day the 2010 notice took effect.62
A short history: 1946 to today
The current list is the third in a sequence of postwar Cabinet Notices. Each one replaced the previous one outright. The arc runs 1946 → 1981 → 2010, with a 2020 elementary-grade redistribution that did not touch the 2,136 total. For the longer arc before 1946, see The History of Kanji: From Oracle Bones to the Jōyō List.
| Year | Instrument | 字種 total | Posture |
|---|---|---|---|
| 1946 | 内閣告示第32号 (当用漢字) | 1,850 | Prescriptive (scope of use) |
| 1981 | 内閣告示第1号 (常用漢字) | 1,945 | Advisory (目安) |
| 2010 | 内閣告示第2号 (改定常用漢字) | 2,136 | Advisory (目安) |
| 2020 | MEXT (education ministry) 学習指導要領 | 2,136 (no change) | Kyōiku redistribution only |
1946: 当用漢字 (tōyō kanji), 1,850 characters, prescriptive
The first postwar list was issued as 内閣告示第32号 (Cabinet Notice No. 32) on 16 November 1946, with a total of 1,850 字.95 The まえがき (preface) framed the list as defining 「使用範囲」 (the scope of use) for laws, official documents, publications, and newspapers. In other words, it was meant as both the minimum and the maximum.95
The reform was drafted by the 国語審議会 (National Language Council) during the Allied occupation. Some SCAP officials saw kanji complexity as a barrier to literacy and democratization, but a 1948 Ministry of Education literacy survey reported only about 2.1% functional illiteracy, undercutting the strongest reformist position.510
Two follow-on tables fixed the shapes and readings: the 当用漢字音訓表 (1948) and the 当用漢字字体表 (1949). Together they fixed the modern simplified shapes (新字体) for the listed set.910 The 当用 table was abolished when the 1981 notice took effect.92
1981: 当用 to 常用, 1,945 characters, advisory
The 1981 notice was issued as 内閣告示第1号 (Cabinet Notice No. 1) on 1 October 1981, with a total of 1,945 字種 and 4,087 readings (2,187 on'yomi + 1,900 kun'yomi).92 The change from 1946 was a net +95 字, with no deletions. The shape change of note was the simplification of 燈 to 灯.2
The framing word switched from 「使用範囲」 (scope of use) to 「目安」 (guideline). The 1981 preface positions the list as 「目安を示す」 (showing a guideline) rather than a binding scope.92 This was an explicit concession to actual usage: people had always used more kanji than the 1946 list allowed.210
2010: revision to 2,136
The 2010 revision was issued as 平成22年内閣告示第2号 (Cabinet Notice No. 2 of 2010) on 30 November 2010, raising the total to 2,136 字種.16 The change from 1981 was 196 added and 5 deleted (勺, 銑, 脹, 錘, 匁), for a net +191.27
Representative additions include 俺, 鬱, 彙, 拶, 籠, 葛, 餅. The revision also added the full prefecture-name set: 茨, 媛, 岡, 潟, 岐, 熊, 香, 佐, 埼, 崎, 滋, 鹿, 縄, 井, 沖, 栃, 奈, 梨, 阪, 阜.27
The 答申 (council report) cited 「情報機器の普及」 (the spread of information equipment) as the rationale. Keyboard input was now part of the literacy environment, so the listed kanji had to be readable. But not every listed kanji had to be hand-writable.7 The same notice altered readings on 33 existing characters, counting additions, deletions, and modifications together. Representative added readings include 関(かか)わる, 応(こた)える, 創(つく)る, 育(はぐく)む, 委(ゆだ)ねる, 旬(シュン).2
The 1981 内閣告示第1号 was abolished at the same time, completing the pattern: each revision replaces the previous notice rather than amending it.16
2020: elementary-grade redistribution (no jōyō count change)
The 平成29年告示 小学校学習指導要領 (2017 MEXT Course of Study notice for elementary schools) took effect on 1 April 2020.811 It added 20 prefecture-name kanji at grade 4. This raised the 学年別漢字配当表 (Grade-by-Grade Kanji Allocation Table) total from 1,006 to 1,026.811
The 20 characters are the same prefecture set that joined the 常用 list in 2010: 茨, 媛, 岡, 潟, 岐, 熊, 香, 佐, 埼, 崎, 滋, 鹿, 縄, 井, 沖, 栃, 奈, 梨, 阪, 阜.811 They were already inside the 2,136 字種 set. The 2020 change moved them earlier within the curriculum, not into the 常用 list.28
The current per-grade kyōiku distribution after the redistribution is:
| Grade | Kyōiku kanji added |
|---|---|
| 1 | 80 |
| 2 | 160 |
| 3 | 200 |
| 4 | 202 |
| 5 | 193 |
| 6 | 191 |
| Total | 1,026 |
Transitional implementation began on 1 April 2018, with full enforcement on 1 April 2020.8
What "adopting" the jōyō list means in practice
The 常用漢字表 gets its force not from its own advisory text, but from four institutional commitments that adopt it uniformly. Each commitment turns the table into a different concrete rule.
Newspapers and broadcasters
Both the 1981 and 2010 prefaces explicitly name 新聞 (newspapers) and 放送 (broadcasting) as expected adopters of the 目安.36 Major newspapers, through the 日本新聞協会 (Japan Newspaper Publishers and Editors Association), and NHK maintain their own house tables aligned with 常用漢字 by default. Non-jōyō characters are either rewritten in kana, paraphrased, or printed with furigana.12
NHK's reference table is consolidated in the 『NHK漢字表記辞典』, edited by the 放送文化研究所 (Broadcasting Culture Research Institute). It is described as 「『常用漢字表』に完全対応」 (fully aligned with the 常用漢字表), while documenting cases of divergence such as homophone disambiguation and kana preference for high-frequency words.12
When a newspaper writes 「拉致 (らち)」 with furigana, it signals that 拉 is jōyō but rare enough to warrant a reading aid. When it writes the verb 「もくろむ」 in kana instead of 目論む, it signals that 論 (jōyō) plus 目 reads fine, but the compound 目論 is outside the house convention.12
Government documents and law
The 常用漢字表 itself is a 目安 (guideline), not a law.4 A separate cabinet-bureau rule is binding for legal drafting.
When the table was revised in 2010, the 内閣法制局 (Cabinet Legislation Bureau) issued an updated 「法令における漢字使用等について」 that translates the table into rules for statute drafting from 2011 onward.4 As a result, the table is advisory in form but mandatory in practice for 法令 (laws) and 公用文書 (official documents), through the cabinet-bureau rule that sits on top of it.4
Schools (the kyōiku subset)
The 教育漢字 (kyōiku kanji) are the 1,026 characters of the 学年別漢字配当表 (Grade-by-Grade Kanji Allocation Table). They are mandatory through grade 6 of elementary school.811 The remaining ~1,110 jōyō kanji (2,136 − 1,026) are taught across 中学校 (junior high school, grades 7 to 9) and 高等学校 (senior high school, grades 10 to 12).211
The current 中学校学習指導要領 (junior high school Course of Study) specifies the sequence: grade 7 students read 250 to 300 of the secondary set, grade 8 add 300 to 350 more, and grade 9 cover most of the rest.211 A Japanese high-school graduate is therefore expected to read (not necessarily write) every character in the 2,136-字種 set.2
What the list does not bind
The table explicitly excludes 「科学，技術，芸術その他の各種専門分野や個々人の表記」 (specialized fields and individual writing).3 Proper nouns are not in scope. The prefecture-name kanji are treated as a special accommodation, not a relaxation of the rule.3
Novels, manga, surnames, place names, technical writing, and personal correspondence can exceed jōyō. The table governs institutions, not the language.34
The jōyō, jinmeiyō, and hyōgaiji boundary
The 2,136-character list sits at the center of three concentric sets. Around it is the 人名用漢字 (jinmeiyō) extension for personal names. Around that is the residual 表外字 (hyōgaiji) category that covers everything else.
Jinmeiyō kanji: the legal-for-names extension
The 戸籍法 (Family Register Law) and the 戸籍法施行規則 (Family Register Law Enforcement Regulations) specify which characters may appear in a child's given name on the family register. The 漢字の表 is published as 別表第二 of the regulation and administered by the 法務省 (Ministry of Justice).1314
The permitted name set is the union of two lists: (a) all 2,136 常用漢字, and (b) the supplementary 人名用漢字 (jinmeiyō) list. The supplementary list contains 863 字種. That figure includes the September 2017 addition of 渾 following a court-driven petition.1514 The combined name-eligible set is therefore 2,136 + 863 = 2,999 字種.15
The supplementary list is revised periodically by ministerial ordinance, typically in response to citizen petitions and court rulings.1315 Many name readings outside the standard on/kun rules are nanori. That is a separate issue at the reading layer rather than the character layer.
Hyōgaiji: everything else
表外字 ("outside the table") names any kanji not in the 常用漢字表. It is a leftover category, not a curated list.2 The set includes classical and literary forms, surnames, place names, technical vocabulary, and Chinese loanwords beyond common use.2
For a sense of scale, the 大漢和辞典 (Morohashi) contains roughly 50,000 entries. This is the standard scholarly reference and an order-of-magnitude indicator of the maximal kanji inventory, not a count of "real" hyōgaiji in modern use.16 Unicode's CJK Unified Ideographs block plus its extensions hold over 90,000 code points. Most are encoded for completeness rather than active use.2
Why a Japanese adult still reads characters they were never taught
The 2,136 set is a literacy floor for institutional Japanese, not a ceiling on adult reading.2 Surnames and place names are the largest source of post-school kanji acquisition. The jinmeiyō list captures part of that, but family-register history preserves variant forms the modern list does not.1314
Adult readers also encounter literary 表外字 through fiction, classical citations, and 漢文-derived idioms (idioms from Literary Chinese).2 People learn these characters from context rather than a curriculum. That is why even Japanese readers occasionally rely on furigana for an unfamiliar surname.
What this means for a learner
The list answers a membership question, not a sequencing question. If you treat it as a study syllabus, you risk confusing what to learn with how to learn it.
A reasonable recognition target, not a finish line
Recognizing all 2,136 jōyō characters brings you close to the literacy floor of an institutional Japanese reader. It covers most of a modern newspaper, government notice, or general-interest book.3 For realistic coverage numbers by corpus, see How Many Kanji Do You Need? A Realistic Count.
The list still excludes surnames, place names outside the prefecture set, technical vocabulary, and literary 表外字, so even full jōyō coverage is not full adult-reading coverage.32
Order: not the same question as "what is on the list"
The 常用漢字表 defines membership and assigns approved readings. It does not prescribe a learning order for non-school learners.3 The 学年別漢字配当表 (kyōiku list) imposes an order on the first 1,026 characters for school use. The remaining ~1,110 are unordered within the 常用 set.811
Frequency-based orders (corpus-derived) and component-based orders (radical-first or Heisig-style) are independent of the 常用漢字表. The table neither endorses them nor forbids them. For the trade-offs, see Should You Learn Kanji in Frequency Order, School Order, or Pedagogical Order?.2
Good to know
The list is a 目安 (guideline), not a law
The Cabinet Notice frames the table as 「漢字使用の目安」 (a guide for kanji use). The 常用漢字表 is a 内閣告示 (Cabinet Notice), not a 法律 (statute) or 政令 (cabinet order). It has no penalty clause.4
Practical force comes from the 内閣法制局 (Cabinet Legislation Bureau) follow-up rule 「法令における漢字使用等について」, which binds statute drafting on top of the advisory table.4 Institutions adopt the table voluntarily and uniformly. That is what makes the list look mandatory in practice.
Why the 2010 revision added kanji that are hard to write
The 答申 rationale was 「情報機器の使用が一般化・日常化している現在の文字生活」: in a world where information equipment is common and used daily, the literacy floor can include characters that are hard to handwrite but trivial to type.7 The standard example is 鬱 (29 strokes). A routine handwritten form is impractical, but the IME supplies it from うつ on demand.7 The implication is that the 2010 list is first a recognition standard and only second a production standard.
What the 33 reading changes did
The 2010 notice altered readings on existing characters in 33 instances, counting additions, removals, and modifications together.2 Representative added readings on existing characters include 関(かか)わる, 応(こた)える, 創(つく)る, 育(はぐく)む, 委(ゆだ)ねる, and 旬(シュン).2 These changes are easy to overlook. The 字種 total gets the headline number, while readings end up as footnotes in most summaries.
Romanization: jōyō vs joyo vs jouyou
「じょうよう」 in Modified Hepburn is jōyō, with long vowels marked by macrons.2 The form joyo is the stripped spelling used when macrons are unavailable. It is the same word, not a separate transliteration.
The form jouyou is the wāpuro (word-processor) input string, not a romanization scheme at all.2 The Japanese government's MOFA (Ministry of Foreign Affairs) passport rules use 「ジョウヨウ」 → Joyo, dropping the macron. That is a registration convention for proper names, not a phonetic spelling.
The set the JLPT does not officially track
The JLPT (Japanese-Language Proficiency Test) publishes can-do statements but no official kanji list at any level.2 The common assertion that "N1 covers all jōyō" is community convention, useful for goal-setting but not a JLPT specification.
See also
- How to Learn Kanji: A Strategic Overview of Heisig, WaniKani, and Kanji-in-Context
- Top 50 Kanji Radicals by Frequency: The 70% Coverage List for Jōyō Kanji
- Writing Kanji by Hand: Is It Still Worth It?
- How to Count Kanji Strokes (画数): The Eight Basic Strokes Plus the Corner, Hook, and Enclosure Rules