改革字 Reformed Chinese Characters

A unifying in-the-middle alternative to Simplified & Traditional Chinese that combines the best of both sets

Quick Links

• 改革字縂表 Reformed Chinese Characters List: 832 Reforms / 3700 漢字 Hanzi
• 改革字例句 500 Example Sentences in order of Reformed, Traditional, Simplified

Introduction

If you’re familiar with written Chinese, you most likely know about the endless, heated debate between Simplified Chinese and Traditional Chinese, both of which have their own pros and cons. Simplified reduces character strokes often at the cost of beauty while Traditional better preserves Chinese culture and aesthetics but may be a hassle to handwrite. In today’s digital age, handwriting is not as much of a high priority versus in the past because most people type and typing Traditional takes the same amount of effort as typing Simplified. For example, pinyin inputting “ma”, just two letters, yields both 馬 and its Simplified form 马 depending on the preferred character set. However when the displayed text size is small, certain complex Traditional characters such as 舊、體、寶 may appear illegible compared to their Simplified counterparts 旧、体、宝. At the same time, Simplified set contains too many visually unpleasing characters e.g. 见、专、风 far off from their historical orthodox forms and it’d be a deep cultural loss to completely discard Traditional for Simplified. Therefore after much meticulous research, I am introducing a new character set, 改革字 Reformed Chinese, to capture the best of both worlds: Traditional’s beauty and Simplified’s legibility.

Important Usage Notes

• The provided Reformed Chinese characters list contains a total of 3700 common characters including 832 Reforms. It is a non-exhaustive template however 3700 characters cover over 99% of the characters found in modern Chinese. Characters outside the list are extremely rare and Reformed still applies where applicable.
• Reformed applies primarily to printed texts, users are free to handwrite whichever they prefer.
• Surnames are always displayed in Traditional out of respect regardless of the characters’ corresponding Reforms. When not used in surname contexts, characters normally appear in Reformed.
• Traditional may still be printed for cultural or aesthetic purposes, Reformed does not obsolete Traditional.
• Reformed is also designed to be a standardized set across the Sinosphere (漢字文化圈) where Chinese characters are commonly used including in Japan. Therefore I have factored in the Japanese language too when compiling these Reforms since characters are not always homophones outside Mandarin. Additionally, 糸 radical like in 統、純、紙、etc always appears as unhooked 糸 instead of 糹closely similar to what’s found in Kangxi Dictionary (康熙字典) and modern Japanese, Korean printing typefaces. Japanese speakers are more than welcome to adopt Reformed too and especially so because Shinjitai (新字体), Japan’s take on character simplification, was a major inspiration. Other influences include 1935 initial simplifications (第一批簡体字), 1969 Singapore simplifications (502簡体字), and 1977 second round simplifications (二簡字), all of which were later officially retracted.

Character Reform Process

Here I’ll detail my general steps when Reforming. There are several factors and guidelines I followed throughout the process: overlap (e.g. 会、点、国 in both Simplified and Shinjitai), resemblance to Traditional (e.g. 變→変、齊→斉), historicity (e.g. 俻 is a variant recorded in Liang dynasty dictionary Yupian 玉篇), consistency (e.g. 儈→侩、澮→浍、檜→桧、etc), logic (e.g. 心 “heart” in 愛 “love”), frequency (e.g. 个、几、从), no oddly regularized cursive (nothing like 东、发、図), no cluttering (e.g. 鑿→凿、釁→衅), no drastic component omissions (nothing like 广、产、乡), sound in other topolects (方言) and languages beyond just Mandarin.

1. Collect the most commonly used Chinese characters. Learn Chinese Characters provides a list of 1500 most common characters ordered by frequency. I used Google Translate to convert the list’s Simplified characters into Traditional. I found another list, DigMandarin’s, which contains 3500 common characters mostly in Traditional but not ordered by frequency. I then combined the two lists and removed duplicates. As Traditional standards vary slightly by region, I adhered to Taiwan standard and replaced characters outside it like 爲、裏、綫 with their respective standard forms (正字). Learn Chinese Characters’ 1500 characters appear first in Reformed characters list followed by DigMandarin’s and other characters I added in to better illustrate as a template, totaling 3700 characters.
2. Compare characters across Traditional, Simplified, Japanese following some adjustments and pick the overlap. In Simplified set, revert all Simplified components to their directly corresponding Traditional forms e.g. 讠→訁、钅→釒、贝→貝. Japanese set mainly refers to Joyo kanji (常用漢字), 2136 commonly used characters including Shinjitai. As Shinjitai in standard usage doesn’t apply to all characters unlike Simplified, extend Shinjitai to non-Joyo kanji too e.g. 摑→掴、疇→畴. For both Simplified and Shinjitai, if the character’s top component is missing a dot which Traditional has, add the dot too e.g. 厅 (廳)→庁、写 (寫)→㝍. Now between these 3 sets, pick the characters where at least 2 sets overlap. For example, Reformed contains 来、医、学 because Simplified and Japanese overlap. Likewise, Reformed contains 愛、後、開 because Traditional and Japanese overlap. Reformed also contains 説、鉄、撹 after said adjustments.
3. If all 3 sets differ, pick the Simplified or Shinjitai that’s closer to Traditional. For example, Shinjitai 剣、焼、関 resemble Traditional 劍、燒、關 more than their Simplified counterparts 剑、烧、关. If both Simplified and Shinjitai are unsatisfactory, select a historical variant (異體字). In very rare cases where Simplified, Shinjitai, historical variants are all unsatisfactory, stick with Traditional e.g. 戶、亞、龍. Fewer character strokes do not always equate to better legibility e.g. Traditional 亞 (8 strokes) is clearer than Shinjitai 亜 (7 strokes). However a Traditional character with both Simplified and Shinjitai forms usually indicates the need for a Reform of sorts. Sometimes Reformed is even simpler than Simplified e.g. 貳 (贰)→弍、鱗 (鳞)→魿.
4. If the overlapping character is still complex, pick a simpler historical variant for clearer legibility e.g. 麼→庅、農→莀、壇→𡊨. Ministry of Education’s Dictionary of Chinese Character Variants (教育部異體字字典) provides an outstanding collection of recorded historical variants that I mostly selected from and also indicates the standard Traditional forms of characters. I additionally consulted 簡化字bot (@jianhuazi_bot) and 新字体bot (@shinjitai_bot) for further historical usages.
5. Keep aesthetic, non-cursive Simplified characters that are too frequently used e.g. 個→个、麗→丽. Many of these are also historical variants. Some common Simplified characters like 过 (過)、时 (時) are not included because otherwise these same Reform logics should apply to all their related characters too for consistency e.g. 鍋、渦、待、特. Also keep aesthetic, non-cursive Simplified characters that return to an earlier form e.g. 雲→云、鞏→巩. However, there cannot be any confusion with earlier form characters still in use today.
6. Merge characters with both similar meanings and same pronunciations including tone e.g. 兇→凶、唸→念. I consulted Wiktionary to ensure characters are homophones in Cantonese and On’yomi (音読み), Japanese readings derived from Chinese pronunciations at the time of introduction, too. I applied overlap mergings found in both Simplified and Japanese usages e.g. 颱→台、讚→賛 as well as historical substitutions e.g. 靈→灵、飆→猋. Merging prerequisites do not apply to extremely obscure characters with no modern usage e.g. 認→訒、嬸→妉.
7. Replace complex sound components with simpler characters and maintain consistency. Sound components (聲旁) traditionally hint at character pronunciations and usually do not provide meaning e.g. 巴 in 吧、里 in 理. Many pronunciations have changed over time so might not make sense in Mandarin today but still might in other Chinese topolects or Japanese. Reformed replaces complex sound components with simpler approximate homophones e.g. 檳→梹、鑹→鋛 or non-cursive abbreviated characters e.g. 傳→伝、還→还. I checked Wiktionary to see whether characters are near homophones outside Mandarin too. Wiktionary was also helpful in discovering nonofficial shorthands e.g. 蔔→苩 (Cantonese)、羹→焿 (Taiwan). I applied many of the same component replacements found in both Simplified and Shinjitai e.g. 機→机、擇→択. Sound component replacements apply consistently across all related characters e.g. 遠→远、園→园、轅→䡇、etc. While sound components often don’t provide meaning, they shouldn’t contradict their characters’ definitions either. I frequently consulted Pleco add-on The Outlier Linguistics Dictionary of Chinese Characters (outlierlinguistics) to distinguish which components are sound as well as find out the original meanings of characters.
8. If all aforementioned options are unfeasible, create a new character based on other Reforms. Many of these new characters do not have Unicode support so I’ve included their character compositions in Reforms list. Simplified, Shinjitai, historical variants inspired these new characters so they are still visually similar e.g. 發→⿱癶矢、臨→⿰臣攵. I also replaced potentially difficult to remember components with simpler, more familiar ones e.g. 滷→⿰氵卤、假→⿰亻㞋.
9. Reduce components consisting of 3 or more reduplicated characters (疊字) e.g. 撬→㧌、疊→畳. Such reductions have occurred throughout Chinese history e.g. 靁→雷、纍→累. Reduction never applies to the base reduplicated character itself e.g. 品、晶、磊 but only when a component in other characters. If the would-be resulting reduction conflicts with existing characters in use e.g. 蕊 and 芯、灄 and 洱, then reference variants instead and apply to all related characters e.g. 鑷→銸、躡→踂. In a couple cases I’ve merged the entire character into the reduplicated component itself if the latter does not appear as a character in modern usage e.g. 協→劦、飆→猋.
10. Apply Reforms consistently to all related characters e.g. 僉→㑒、檢→検、撿→𢮦、etc. Do not Reform related characters differently unless Simplified and Shinjitai overlap following mentioned adjustments e.g. 燈→灯、證→証. To ensure consistency, I pulled all related characters from Chinese Text Project Dictionary and compared with what’s in Reforms list to avoid missing any characters.
11. Optimize stroke count balance in compound characters and sentences. For example, compounds 寶貝、藝人、職工 have poor balance compared to 宝貝、萟人、聀工 due to having a complex character first followed by a much simpler one. The ideal balance is that all adjacent characters share approximately the same complexity. I experimented by converting Chinese sentences gathered from social media into Reformed, placing them next to Traditional and Simplified for comparison. I often repeated earlier Reform steps when characters still looked unbalanced together.

Reformed Characters List

I aimed to keep Reformed Chinese characters list straightforward and self-explanatory. You are highly encouraged to comment, critique, ask questions anywhere in the Google Sheets, please indicate which character set you normally use. I’d love to know your thoughts e.g. you hate a Reform, wonder why a Reform is such a way, believe a character should stay in Traditional, think this project is cool, etc. The leftmost column displays 3700 common Traditional characters following Taiwan standard. You will notice the Chinese name for “Traditional” in column header is「傳統字」instead of the more well-known name「繁體字」. This is because Simplified and Traditional users each have different names for the character sets. Simplified users commonly call Traditional「繁体字」“complex form characters” which is mostly a misrepresentative umbrella term with a biased, often negative undertone. In contrast, many Traditional users call Traditional「正體字」“standard form characters” which isn’t accurate when encompassing the entire Sinosphere where standards now differ regionally. Therefore I literally translated the English name “Traditional characters” into「傳統字」which in Reformed would be「伝統字」, 統 having unhooked 糸 instead of 糹. The next column contains 832 character Reforms. Blank row means no change from Traditional otherwise I listed the Reform including character composition if no Unicode support. The following column contains justifications, usage notes, and Unicode appearance info. The header reminds you to filter columns for easier, selective viewing. Maybe you’ll want to see only character Reforms or ones without proper Unicode support. The last column is Unicode support, blank row means there already is correct Unicode support. When unsure whether a Reform has Unicode support I searched in Chinese Text Project Dictionary e.g. “氵+食” for 湌、“言+上” for “No matching characters found”. At least 249 characters might display different Unicode appearances across other character sets and Reformed’s purpose is to unify all into one standard. For example, Reformed 亮 should have 儿 at bottom as found in Traditional Chinese and Japanese Unicodes instead of 几. 93 Reforms have Unicode data but no proper support so they normally will not display e.g. 𧥾 (⿰言戈)、𮬥 (⿰鳥几), however you can still copy and paste them into GlyphWiki search bar for an idea of how they appear. I also included GlyphWiki links for many Reforms. 203 Reforms do not have any Unicode data or support. In previous Notes column I detailed how Reformed Unicode should appear, these are the main ones which currently have Unicode appearance issues due to merging:

• ⻌ never ⻎
• Unhooked Chinese Unicode 糸 when left-side radical never 糹

Left-side radical 糸 like in 縂

• 言 top component is always 亠 never 二
• 艹 is always one connecting 一 stroke
• ⻀ is always one connecting 一 stroke
• 今 is always ⿱亼㇇ (Hong Kong, Taiwan, Japan, Korea, Vietnam standard)
• 令 is always ⿱亼龴 (Hong Kong, Taiwan, Vietnam standard)
• 戶 never 户/戸
• 月 never ⺼
• 雨 as top component is always ⻗
• 戋 never 㦮
• 林 never 𣏟
• 匀 never 勻
• 单 never 単
• 圣 never ⿱ス工 (e.g. 𦙾 not 胫)
• 亦 as top component is always 亦 itself
• 术 never 朮
• 与 final 一 stroke does not extend across ㇉
• 呂 never 吕
• 奥 never 奧
• 彐 never 彑 except in 互 derivatives
• 录 never 彔
• 尔 never 尓
• 昷 never 𥁕
• 木 never unhooked 朩
• 負 never ⿱刀貝
• 内 never 內
• 彦 never 彥
• 産 never 產
• 黄 never 黃 except when surname
• 吴 never 吳 except when surname
• 步 never 歩
• 堇 top component is always 廿 never 艹
• 𦰩 top component is always 廿 never 艹
• 䍃 top component is 夕丶like in 然 never 爫

Example Sentences

I gathered 500 innocuous example sentences to demonstrate how Reformed looks in practice, placing them next to Traditional and Simplified for comparison. Only characters with Unicode support appear however you are free to reference Reformed characters list and try yourself converting Chinese sentences into Reformed. Do note Reformed’s emphasis on both beauty and legibility.

Potential Applications

As mentioned, Reformed’s primary purpose is to unify Simplified and Traditional into one same standard and potentially Japanese kanji too. Most Simplified users are reluctant to adopt Traditional and same vice versa. Sharing the same script may help bridge the gap in bringing people closer together. Simplified and Traditional have unfortunately become too politicized nowadays so Reformed offers a clean, new slate. Reformed as a standardized set for the Sinosphere would also simplify Unicode mergings in Han unification which now has to account for all the different regional varieties. Alternatively, Reformed can serve as a stepping stone for Simplified users in case they choose to re-adopt Traditional in the future given the current resurgence and love for traditional Chinese culture. At the very least Reformed is closer to Chinese historical orthodoxy and also practical for calligraphy (書法) while still being legible. Let’s preserve the beauty of Chinese characters and culture forever.

Acknowledgements

Thank you very much

• 1500 Most Common Chinese Characters, Learn Chinese Characters
• 3500 Commonly Used Modern Chinese Characters List, DigMandarin
• Chinese Text Project 中國哲學書電子化計劃
• Dictionary of Chinese Character Variants 教育部異體字字典
• GlyphWiki 字形維基
• Kingdom of Characters: The Language Revolution That Made China Modern, Jing Tsu
• Long Story of Short Forms: Simplified Chinese Characters from A to Z, Roar Bökset
• Outlier Linguistics Dictionary of Chinese Characters
• Pleco Software
• Wiktionary 維基詞典
• 簡化字bot (@jianhuazi_bot)
• 新字体bot (@shinjitai_bot)

Contact

Please remember you are encouraged to comment, critique, ask questions about Reformed Chinese characters list in the Google Sheets itself, I will address them there. Otherwise you may email me at ReformedChinese@gmail.com. You are also welcome to follow @ReformedChinese on Instagram. Thank you!

―盧豫楷 Zeo Lu