When it comes to gairaigo, every time there's a consonant cluster or a final one you generally add a -u to it. Glass=gurasu, plan=puran, story=sutoorii, wrestling=resuringu. As far as I can tell, this is the case for everything except the clusters /tr/ and /dr/, and final /t/ and /d/. Dragon=DOragon, dress=DOresu, trap=TOrappu, trick=TOrikku, apart(ment)=apaaTO, fight=faiTO. What I'm guessing is that "tsurikku" would have one too many consonants, the analogue "tsrick" doesn't sound great. Final ツ is a lot less popular but seems to exist, when the word ends in /ts/ or /ds/. And of course when it'd be /dzu/, it becomes /zu/ because of them being the same in Tokyo-ben, therefore good=guddo but goods=guzzu.

I'm looking for papers that investigate this, it looks pretty interesting.

  • Flute is フルート but fruit is フルーツ for some reason. Not sure if it came from the plural form for counting kinds.
    – aguijonazo
    Commented Jan 22, 2022 at 14:16
  • @aguijonazo Yeah, and shirt is シャツ. It doesn't even have the long vowel you'd expect. There are definitely some irregularities.
    – user48723
    Commented Jan 22, 2022 at 14:24
  • 1
    Maybe not exactly what you are looking for, but this seems to discuss distributions of sounds in loanwords.
    – sundowner
    Commented Jan 22, 2022 at 22:26
  • 2
    Another. I haven't looked at it carefully enough, but this (and cited artilces) might be more relevant.
    – sundowner
    Commented Jan 23, 2022 at 5:24
  • 1
    @sundowner Right off the gate, it looks like exactly what I'm looking for. Awesome, thanks. Let's post this as the answer.
    – user48723
    Commented Jan 23, 2022 at 15:13

2 Answers 2


According to the following article, -u was the most common vowel and there has been change to -o. It further rationalizes the change using some theory.

Copying some relevant(-looking) passages and citations for later reference.

挿入母音U [ɯ] は最も出現頻度の高いことから「デフォルト」の挿入母音と呼ばれることがある。 U が最もよ く挿入されるのは、日本語の 5 母音の中で①内在時間長が最も短く、②聞こえ度が最も低く、③無声化が起こり やすいことから、最も母音が無いように聞こえるためである (窪薗1999:229-233) とされる。

ここで上述の環境で母音挿入の変化が起きた要因を考察する。 まず、 ʦU から tO への変化 (10a) については、 初めはデフォルト挿入母音 U が支配的だったが、後に原語の子音/t/の閉鎖音としての素性の保持を優先するよう になり、 O が挿入されるようになったと説明できる。

  • 窪薗晴夫 1999『日本語の音声』東京:岩波書店
  • Kubozono, Haruo. 2002. Prosodic structure of loanwords in Japanese: syllable structure, accent and morphology. Journal of the Phonetic Society of Japan 6(1). 79-97.
  • Lovins, Julie B. 1975. Loanwords and the phonological structure of Japanese. Bloomington: Indiana University Linguistic Club
  • 澤田田津子 1985「外来語における母音添加について」『国語学』 143. 88-75.

I think you are right. This should be simply because Japanese had no sound that was similar enough to "tu" and "du". As you know, phonetically speaking, the タ-line is highly irregular:

タ /ta/ ダ /da/
/t͡ɕi/ /d͡ʑi/
/t͡su/ /zu/ (/d͡zu/)
テ /te/ デ /de/
ト /to/ ド /do/

Although Japanese people started to recognize トゥ (/tu/) and ドゥ (/du/) relatively recently as "extended kana", these were not things available in the past. If old Japanese people had added "u" unconditionally, trap would have ended up with ツラップ, but /t/ and /t͡su/ must have sounded too different even to their ears. So they had to use ト instead.

At the end of words, there are some irregularities like シャツ, バケツ, スーツ and ビーズ. Perhaps people did not pay much attention to this because sentence-end vowels were sometimes simplified even in ordinary Japanese.

  • About the word-final instances of ツ and ズ, I wonder how much of that might have been influenced by the English plurals? Commented Jan 24, 2022 at 18:09

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .