Convert Kana to Hentaigana Program 正体仮名を変体仮名に転換できるプログラム

Disclaimer 記事を読む前の重要事項
This program is now deprecated and likely will not work properly. Please read about my updated hentaigana input software here.

このプログラムは現在、利用停止されていて、上手く機能しないだろう。更新した変体仮名インプットメソッドについては、これを読んでください。


When I was in Hokkaidō, I visited the Sapporo Beer Museum. I was impressed by the quality of the beer and the information provided during the tour. But as I was leaving, I saw this.

北海道に行って、札幌ビール博物館を訪れた。ビールの質とツアーで聞いた情報が印象的だと思った。しかし、去っている時、これを見た。

barrels.jpg

I immediately got excited as I realized this was hentaigana, kana characters that became nonstandard after the 1900 script reform. Despite how they are quite rare now, they were very widely used prior to the Meiji period. So how do we read it? Fortunately, there was a translation too.

これが変体仮名(1900綴り革新の後、標準仮名ではなくなった仮名)だと悟り興奮した。最近では珍しいけれど、明治時代以前に広範だった。とにかく、読み方は何だろう?幸い、翻訳もあった。

barrels translation

But let’s say we wanted to write this in a word document. Previously, we would have to save each character as an image, but after the Unicode 10.0 update in June 2017, 287 hentaigana characters are included in Unicode.

仮に、マイクロソフトワードでこの句を書きたい、としよう。以前は、各字をイメージとして保存することが必要だったが、2017六月のユニコード10.0アップデートの後、ユニコードが287変体仮名の文字を含むようになった。

This is all well and good, but how do we actually write these characters. Previously, the easiest solution would probably be to head to Wikipedia (https://ja.wikipedia.org/wiki/%E5%A4%89%E4%BD%93%E4%BB%AE%E5%90%8D) and copy the character you want. But recently, I developed a lightweight program that streamlines this problem. It doesn’t really have a flashy name; I just call it Convert Kana to Hentaigana. To use it, you type the standard kana of the hentaigana you want, and then a list of available hentaigana will appear on the right-hand side. Then, you press the number or symbol of the hentaigana you want, and the program will replace the standard kana with that hentaigana. Here is an example of me typing the sentence on the barrels.

変体仮名を含んでいるのはいいけれど、どうこれらの変体仮名を書こう?以前は、一番簡単な方法はウィキペディアに行って(https://ja.wikipedia.org/wiki/%E5%A4%89%E4%BD%93%E4%BB%AE%E5%90%8D)、変体仮名をコピーすることだ。しかし最近、私はこの問題を解決するための軽量プログラムを作った。カッコイイ名前じゃない、正体仮名を変体仮名に変えよう!(Convert Kana to Hentaigana)という名前だ。このプログラムの使い方は、転換したい変体仮名の正体仮名を入力して、右側に使用可能な変体仮名の表が出る。そして、転換したい変体仮名の数、または記号を押すと、プログラムは正体仮名を変体仮名に換えることができる。ここに例として、私は写真に書かれたの文を入力した。

Barrels type example

The font the program uses is called UniHentaiKana (it’s also used on this site). You can download it from https://wakufactory.jp/densho/font/hentai/. The program can run without the font installed, but you won’t be able to display the text in any other programs, so it’s highly recommended.

プログラムが使うフォントはUniHentaiKanaと呼ばれた(また、このサイトで使われている)。https://wakufactory.jp/densho/font/hentai/ からダウンロードできる。このフォントをインストールせず、このプログラムが実行できるけれど、他のプログラムではテキストが表れないので、フォントのインストールを強く勧める。

You also might notice, Kana Only Mode. What this mode does is suppress the Windows Japanese IME after each kana, so that the kanji selection box doesn’t interfere with the hentaigana selection box. Basically, it avoids this.

また、仮名だけモード(Kana Only Mode)にも注目してほしい。このモードは格仮名を入力した後で、漢字の選択のボックス(IMEボックス)が変体仮名の選択のボックスが差し支えないように、ウィンドスの日本語のIMEを使用できないようにした。つまり、この機能が、以下の問題を防ぐ。

cant see list

This mode would be more convenient if you’re only typing in kana. For example, if you were typing the hentaigana iroha poem, it might be more convenient to use Kana Only Mode. (Whenever there is a squiggle underneath the text, there is a Windows Japanese IME selection box; I couldn’t capture it with my screen recorder).

このモードでは、仮名だけで書いた方が便利だろう。例えば、変体仮名いろは歌を入力するとしたら、仮名だけモードの方が便利だろう(テキストの下に波線がある時、ウィンドスの日本語のIME選択ボックスが出ている。スクリーンレコーダーはそのボックスを読み込めない)。

変体いろは47字

kana mode

So, what are you waiting for? Why don’t you digitize some old texts, or just mess around with this program? You can download it from http://www.mediafire.com/file/m2ctjxa4y9w5ycd/ConvertKanaToHentaigana.zip today!

さあ、そこの君、今、暇をしている?だったら、このプログラムを使って、古いテキストをデジタル化しよう!http://www.mediafire.com/file/m2ctjxa4y9w5ycd/ConvertKanaToHentaigana.zip からダウンロードできるよ!

Post a comment

Private comment

No title

Hi, I just came across your website, your website is a hidden gem. Thank you for sharing your interesting insights. By the way, is the ”ゆふ” in "麦とホップを製す連者ビイルとゆふ酒に奈る", correct spelling? Should it not be "いふ" in 歴史的仮名遣い?

Re: いふ vs ゆふ

Thank you for the kind words! They mean a lot to me.

As for your question, the TL;DR is that both いふ・ゆふ spellings existed at that time and neither was necessarily “correct”.



If you want to know more, I will explain why Historical Kana Orthography can have two spellings which were “correct”.

To give a little context, let’s talk about how early-Meiji Japanese (that on the barrels) developed into contemporary Japanese. Before 1900, a Japanese person could use a multitude of different kana to represent one sound. However in the 1900 Spelling Reform, the Meiji government selected one kana for each sound (a principle known as 一音一字), and the rest of the kana became branded as hentaigana 変体仮名. However, hentaigana usage did continue to some extent, and still does now. The next reform was the post-WWII Script Reform. This reform made changes in an attempt to unify the spoken and written forms of Japanese (a principle known as 言文一致). The system they created is known as Modern Kana Orthography 現代仮名遣い. Any system deviant from Modern Kana Orthography would henceforth be branded as Historical Kana Orthography 歴史的仮名遣. But Historical Kana Orthography continues to be used to a limited extent even today. I occasionally read contemporary research papers written in Historical Kana Orthography.

Knowing this, we can conclude that the orthography of the barrel text differs from contemporary Japanese orthography in that it 1) uses hentaigana 2) does not follow Modern Kana Orthography.

I explained these reforms to demonstrate two ideas of pre-reform Japanese. Firstly, before these reforms there was no true notion of “correct” and “incorrect” spelling. Edo scholars did write works on “right” and “wrong” orthographic norms, but this was a time (similar to Early Modern English) where if the reader could understand the text, then the spelling was correct. Secondly, while Modern Kana Orthography is a system which is clearly defined, Historical Kana Orthography was simply any orthography which existed in the past which did not conform to this new standard. Historical Kana Orthography itself is not a monolith. Therefore, since both いう and ゆう could both be understood, one was just as “correct” as the other before these reforms.

(As a quick aside, the same could be said about hentaigana. Before the 1900 Spelling Reform, there was nothing more “hentai” about any kana in comparison to another. The writers were at free liberty to use whatever kana they wanted, as long as the reader would be able to understand it.)



If you still are curious, I will explain how the いう and ゆう spellings came into existence in the first place.

In the Nara period, the word ‘to say’ was phonetically /*ipu/. During the creation of kana in the early Heian period, this /*ipu/ would be written as いふ (the /h/ sound in Japanese was /p/ at this time). However, by the end of the Heian period the /p/ sound in the middle of a word disappeared. This meant that the word ‘to say’ went from being pronounced as /*ipu/ in the early Heian period to /*iu/ in the late Heian period, the same way that it is pronounced by many today. However, in the Kamakura period, the /*iu/ sound became /*yuu/, a phenomenon known as chōon-ka 長音化. From then on, there are attestations of ゆふ in writing. For example, in 土井本周易抄〔1477〕 we can see it in the phrase 「さて又吉となぜにゆわぬぞ」. This created a divide between those who wrote the word ‘to say’ using the more traditional いふ and the more accurate to pronunciation ゆふ (Even though the pronunciation of both was ゆう). A lot of things may impact why a person would write one over the other such as age, content of writing, location, to whom they’re writing, etc. The significance is that both forms existed to some extent. Eventually in the post-WWII Script Reform, the Japanese government decided that the proper way to write ‘to say’ is 言う. However, even today I hear many people read the word as ゆう, and in manga and other less-formal Japanese texts I have seen it written as such.

I hope that these explanations were helpful. Please don’t hesitate to ask me further on this topic. And of course, if you have any questions like this in the future feel free to contact me (responding to this comment is likely the best way to do such).

Happy studying!

No title

Hi, thanks for the detailed explanation!
Let me explain my thoughts in a bit more detailed manner.

What I thought is ゆふ may not completely consistent spelling with phonology, and it looks like an “overfitting”.
In the example 「さて又吉となぜにゆわぬぞ」, ゆわ is the 未然形 form so the 終止形 form would be ゆう (if it were ゆふ, then I think it should be ゆはぬぞ), and this makes sense as it is a phonetic representation of the sound yuu, even though it is different from both the standard 歴史的仮名遣 (or perhaps 契沖仮名遣=partially a theoretical reconstruction on how the “correct” spelling should be based on Heian phonology) or the modern standard spelling. However, to me, ゆふ is like a hybrid and it seems to represent a sound that did not exist during history, especially considering that the ハ行転呼 has occurred before (late Kamakura period) than the 長音化 phenomenon (Kamakura period). Thus, considering that the text was written during the Meiji era, which would mean practically it is the same how modern Japanese would pronounce (yuu or maybe iu), I imagined that the person tried to write in the correct 仮名遣 but rather chose the “wrong” one which “looks” correct.
What would be your thoughts?

Re: いふ vs ゆふ

Hello,
I understand why you would say that ゆふ is a hybrid, but I have a few ideas. First, I would be cautious in saying that Keichū’s kana orthography is a complete reconstruction of Heian phonology. He had a few mistakes, such as positing the spelling あるひは for the word あるいは when the true etymology is ある+い+は. I think that this spelling is a bit more egregious because this word did exist in the Early Heian period and it was written only as あるいは.

I also want to add upon what I said before. I forgot about the one extra step for ゆふ to become its own verb. In the Kamakura period, when 言ふ was pronounced as /*yuu/ it was reanalyzed as a new verb with a yodan 四段 conjugation. Therefore, there was a split between いふ which has the paradigm
未然形:/*iwa/
連用形:/*ii/~/*yuu/~/*iQ/
終止形:/*yuu/
連体形:/*yuu/
已然形:/*ie/
命令形:/*ie/
And ゆふ which has the paradigm,
未然形:/*yuwa/
連用形:/*yui/~/*yuu/~/*yuQ/
終止形:/*yuu/
連体形:/*yuu/
已然形:/*yue/
命令形:/*yue/
This reanalysis is significant because otherwise the forms ゆわ and ゆい would not exist.

Next, about my example 「さて又吉となぜにゆわぬぞ」 this was simply a mistake on my part. You are correct that in this sentence ゆわ is a conjugation of ゆう. The example I was using comes from the Unabridged Japanese Dictionary 日本国語大辞典. The headword for this entry is ゆ・う[ゆふ] 【言・云・謂】. Note that this dictionary provides the Historical Kana Orthography as ゆふ. For the examples of ゆふ the dictionary provides one sentence from each of the following sources: 斯道文庫蔵帝範応安元年点〔1368〕, 土井本周易抄〔1477〕, 玉塵抄〔1563〕, 天正本節用集〔1590〕, 日葡辞書〔1603~04〕, 浄瑠璃・新うすゆき物語〔1741〕, and 童謡・人買船〔1920〕. Save for Nippo Jisho 日葡辞書, these are all very obscure sources. I am quite surprised that more well-known sources are not cited. Also of note is that all of these sources, except for 斯道文庫蔵帝範応安元年点 present the verb as conjugations of ゆう. The except of 斯道文庫蔵帝範応安元年点 is 「其の要言を聚て近き誡め為すと云(ユふ)こと爾」. So I did a little investigation and I found the article 中世東国文書の言語研究 (https://ci.nii.ac.jp/lognavi?name=ir&lang=en&type=pdf&id=http%3A%2F%2Fid.nii.ac.jp%2F1632%2F00025041%2F&naid=120006513338) which specifically discusses how いふ and ゆう existed concurrently (page 17), and one of the examples is 法言(トユフ)は礼法之言(コト)謂(ヲイフ). In this phrase 言 is glossed as ユフ. However, the same line from the Unabridged Japanese Dictionary is presented in this article as 「其の要言を聚て以て近き誡(め)[sic]為すと云(ユ)こと爾」. Therefore, while these texts show that ゆふ was attested to some extent, their obscurity and reliance on transcription make them not the best piece of evidence. I checked a Japanese translation of the Nippo Jisho and the entry was “Yui, yǔ. ユイ、ゥ(言ひ、ふ)話す.これは本来の正しい言い方ではない. Iy, yǔ(言ひ、ふ)の条を見よ.” While this translation was published in 1980, it once again is an instance where ゆふ is written as the Historical Kana Usage. While I do not love any of these attestations, based on the few that I presented, I think it is safe to say that spelling ‘to say’ as ゆふ was not an isolated incident.

But before we talk about what this all means, I think it is necessary to explain why I had such difficulty in finding attestations of both ゆう and ゆふ. I suspect this is due to the type of texts that would use ゆう・ゆふ. Because this word is in essence “slang” it would be more appropriate in correspondences between friends rather than literature/poetry. However, the type of texts that are digitized and easily searchable are those of literature and poetry. In addition to this, this word came into existence when kana-kanji majiribun was becoming the language of the commoners. In the Heian period, when texts by women were written mainly in kana, finding instances of varying pronunciation may have been easier, but in the Kamakura period when authors would write 言ふ for both いふ and ゆふ it is impossible to know which way it was intended to be read, and if the author even had a preference. In addition, there are also the words 夕, 木綿, and 結ふ which all have the reading ゆふ which makes it more difficult to find relevant search results.

So now that we have established that ゆふ was an attested spelling of /yuu/ ‘to say’, even if we don’t have access to too many of those attestations, let’s talk about the theory behind this spelling. The reason why someone in the Kamakura period would presume that the yodan conjugated verb /yuu/ has a spelling of ゆふ is because Japanese phonotactics forbade native Japanese words to have vowel sequences. This means that ゆう /*yuu/ would not be allowed. The only reason they exist now is because of phonetic change. Therefore, any yodan verb which pronunciation-wise ended with -う in the Kamakura period, would have to have been phonetically and orthographically -ふ in Early Heian Japanese. Therefore, verbs such as 思う, 競う, 襲う which would have been read as /omou/, /kisou/, and /osou/ in the Kamakura period, must have had the form 思ふ, 競ふ, and 襲ふ which would have been read as /omopu/, /kisopu/, and /osopu/ in the Early Heian period. Therefore if we were to imagine the phonetic shape of /yuu/ in the Early Heian period, it would have been /yupu/ ゆふ. And in fact, the modern Japanese 結う went through that exact process. It was /yupu/ in the Early Heian period, and /yuu/ in the Late Heian period and /yuu/ in the Kamakura period, where it was homophonous to ゆふ ‘to say’. In Historical Kana Orthography 結う is 結ふ.

So with that said whether the spelling ゆふ is “incorrect” or “correct” is up to the person reading it. It does make the verb seem like it is much older than it actually is, but the alternative would break orthographic tradition, so it was at the writer’s discretion to choose as they deemed fit.
プロフィール

ジェイク・アイアナコーン

ジェイク・アイアナコーンが日本語を勉強しようと思ったきっかけは、2014年に訪れたディズニー・ワールドの世界の文化展示エリアで、日本文化を体験したことです。日本語が分かるようになりたいと思い、自分で平仮名、カタカナを覚え、単語や文法、そして漢字を独学で学びました。そのうちに日本語を勉強することが好きだということに気づきました。2015年頃には古典日本語の記事を読み、とても興味を持ったので、古代日本の言語と文化の勉強も始めました。また『古文友達』というアプリを作りました。将来は日本で古典文学を研究し、その素晴らしさを世界中に広める研究者になりたいです。みなさんにこのブログの記事を読んでもらい、楽しんでもらえたら嬉しいです。また私の記事に関して思ったことなどコメントしていただけると幸いです。

最新記事
月別アーカイブ
カテゴリ
リンク
検索フォーム