Inspired by https://japanese.stackexchange.com/a/38733/3295

Some examples from Wiktionary:

真っ暗 (makkura, "total darkness")
真っ向 (makkō, "directly opposite", "right in front")
真っ黒 (makkuro, "pitch black)
真っ最中 (massaichū, "infull swing, in the midst")
真っ青 (massao, "deep blue")
真っ赤 (makka, "bright red")

Also possibly related: 真ん丸, 真ん中.

The usual rules for gemination don't seem to apply: 真{ま} doesn't end in or , so where does っ come from?

@user4092 puts forward this theory in a comment:

As for 真っ白, some claim that it derives from prefix "ma" with reduplicated adjectives, which is shared with Filipino or Melanesian languages. This feature is well reserved in "ma aka aka" → まっかっか.

To be honest I'm somewhat skeptical of this.

For now my pet theory is that it's related to the reading まこと, but is there a commonly accepted explanation?

  • 1
    Gemination seems to occur in colloquial speech quite a bit. Often it seems to occur in situations where emphasis would otherwise be applied. eg ばかり->ばっかり. I wonder whether something like this is what's happening in these words you list.
    – A.Ellett
    Jun 30 '17 at 18:56
  • I concur with A.Ellett; because of its meaning as an intensifier, it seems natural for ma- to be lexicalized with emphatic gemination, like matakumattaku, yohodoyoppodo, minaminna etc. If it was related to makoto, I'd expect it to have once been mak-, in which case there should be a maki- or maku- reflex, but there seems to be no such thing; in the Portuguese 16th-century materials it only appears as either macoto or maccuro etc. Jun 30 '17 at 19:05
  • For the record, the Kokugo Daijiten describes its origin as 接頭語「ま」の下に促音の挿入された形, presumably for emphasis. There's plenty of non-geminated words with the 真/ ma- prefix (like indeed makoto itself, probably from "true thing" = ma+koto). Jun 30 '17 at 19:15
  • 1
    [真昼]{まひる}なのに[真昼間]{まっぴるま}、[真新]{まあたら}しいのに[真]{ま}っ[新]{さら}、[真四角]{ましかく}、でも[真]{ま}ん[丸]{まる}、[真正面]{ましょうめん}と[真正面]{まっしょうめん}. 真上、真夏は「まうえ」「まなつ」なんですね
    – Chocolate
    Jun 30 '17 at 22:32
  • 1
    – Chocolate
    Jul 1 '17 at 4:21

(Seems like we're not getting anything more specific, so I'll format my comments as an answer for ease of future reference.)

The electronic edition of the Kokugo Dajiten has it like this:


So the standard view isn't that it was makoto > maQ-; but rather a simple ma-, to which -Q- was later tacked on. And indeed, of the words you've cited, most of them have non-geminated attestation in historical documents; though that, in itself, isn't decisive, since gemination wasn't always written down consistently. Still, I think the standard analysis is more likely true than not.

Why? If you consider some prefixes ending with -ki/-ku that create gemination, you'll find these are often on’yomi, i.e. Chinese loans, like 悪- aku- in 悪口 akkō, akku or 石 seki- in 石鹸 sekken. But these trace straightforwardly to Middle Chinese readings ending in consonants, such as *ʔak and *dźak in this example (compare the modern Cantonese readings: ok, sek). We know these loans retained their final consonants even in Japanese for a good while. It's natural that a syllable-final consonant would coalesce into a geminated one

What that has to do with anything? Well, ma- is a native Japanese prefix, seen in the 8th-century Man'yōshū in expressions like ma-kanasimi "truly sad", or 旅とへど真旅になりぬ "I said I'd set out on a [little] journey, but—lo and behold!—now it has become a real [long] one." Linguists have many reasons to believe that Old Japanese had no gemination, so ma- was a perfectly cromulent prefix by itself, predating the introduction of geminated consonants into the language. In fact, it's very likely that makoto itself is just ma+koto 真事/真言, true things/true words (again as per Daijiten).

Since we had a native Japanese ma- rather than a Chinese mak-, there seems to be no reason for it to trigger gemination. True, the Japanese onbin changes also triggered gemination, like toritetotte; but not with [k], since the onbin form of kakite resulted in kaite instead. Then whence the -Q- in maQ-? Consider that both historically and synchronically, Q is added to create emphasis:

  • matashi, mataku (historical form) → mattaku (modern form)
    • → まっっったく (tongue-in-cheek superemphasis in manga or the Internet, etc.)
  • yohodoyoppodo (both forms still coexist)
  • minaminna (also works with nasals)

Consider also that there are several other emphatic prefixes with a -Q- (取っ, 引っ, 打っ, 押っ as in 押っ死ぬ、押っ放り出す…), and also the use of Q in interjections like "こわっ!"

Given the meaning of 真 as "true, truly, for real", it makes sense for it to receive emphatic gemination, which in time lexicalized into forms like 真っ黒, etc. We can predict from this that, if a word has a 真 prefix but didn't get an emphatic っ fossilized into its lexical entry, it could still get a っ for extra discursive emphasis; and in fact, forms like まっこと or 真ん夏 are easy enough to find online.

  • just found this paper; it's a bit over my head but I wonder if it explains some of it. Jul 2 '17 at 17:05
  • Does Q regularly have the allophone -s- before a vowel? Or is 真っ青 an outlier?
    – jogloran
    Jul 3 '17 at 1:35
  • 1
    @jogloran Good question. That would make sense historically at least, because Old Japanese didn't like vowel sequences anyway. But, I can think of no other vowel-starting words getting an /s/ after /Q/, and sao is attested in the Man'yō as 人魂乃佐青有公之 (ひとたまの さを-なる きみが). Daijiten says this sa- is a prefix, presumably the same emotive one seen in さ夜, さ霧, さ迷 etc. At any rate it seems to be the case that さお is an archaic variation (or compound) of あお, preserved in 真っ青; it's common for bound compounds to preserve ancient forms like that (cf. the sound of English break vs. breakfast). Jul 3 '17 at 14:52
  • 1
    @jogloran With the help of a Python script, I found a single other candidate: 真っ新 まっさら < まっ + あら(た) ? . However, the existence of an independent さら=new (used like さらの浴衣) again muddies the waters. Jul 3 '17 at 14:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.