Aduerbia sonus: Ideophones in two 17th century grammars of Japanese

One of my projects here at The Ideophone has been to track down early sources on ideophonic phenomena. For example, I have suggested that we may call the 1850’s the decade of the discovery of ideophones in African linguistics. But we can push back the linguistic discovery of ideophones a little further by looking to other traditions. Today we look at Japanese, for which I have found some early 17th century grammatical treatises that offer information on ideophones (nowadays called ‘mimetics’ in Japanese linguistics). 1

Back then, it was not very clear to Western grammarians that imitative words could imitate anything besides sound, and therefore our first source, Diego Collado’s Ars grammaticae Iaponicae linguae, calls them “adverbia sonus” (it would be interesting to know whether Japanese words for the category itself —like giongo/gitaigo— already existed back then). Here’s an excerpt from Collado 1632:

Aduerbia concludendi, & aduertendi

Aduerbia sonus sunt multiplicia secundum diuersitatem quam Iapones in sonus terminatione percipiunt, & illis, to, solent postponere: v.g. ua ua to xite, vociferando dicentes, ua ua, & si illis postponitur meqi, u, significat talem strepitum facere: v.g. ua meqi, u, va dicendo vociferor, aris, &c.

Adverbs that conclude and claim attention

The adverbs of sound (adverbia sonus) are many and vary in accordance with the way that the Japanese perceive the sound. The particle to is added to them; e.g., va va to xite ‘vociferously saying wa wa,’ and if they add meqi,u, it means to make even a louder noise; e.g., va meqi,u ‘to shout saying wa.’ [transl. by Richard Spear 1975]

According to Kimi Akita, who has kindly helped me to identify these constructions, the particle to can be identified with the quotative particle or complementizer, used in contemporary Japanese to introduce reported speech and adverbial ideophones. Collado’s first example can thus be glossed as follows:

waawaa-to it-te
‘saying waawaa’

What Collado transcribes as ‘meqi,u’ can probably be identified with the verbalizer ‘-meku’ which (according to Kimi Akita) is less productive nowadays. Kimi provides some interesting examples of lexicalized verbs derived from ideophones using this suffix:

(1) mimetic: kira-meku ‘twinkle’ (< kirakira), zawa-meku ‘hum’ (< zawazawa), hira-meku ‘be inspired’ (?< hirahira) (2) nonmimetic (rare): haru-meku ‘get like spring’ (< haru ‘spring’), huru-meku ‘get old’ (< huru-i ‘old’), nazo-meku ‘look mysterious’ (< nazo ‘mystery’) [Kimi Akita p.c.]

Attaching this suffix to a monomoraic root like ‘wa’ is not allowed in Modern Japanese, notes Kimi.

Landresse 1825 [based on Rodrigues 1604]

But there is a fragment that is more interesting and that takes us even further back; it is found in Rodrigues’s Arte da lingoa de Iapam. I have not been able to consult the original and am relying on an abridged French version published by Landresse in 1825. Here is what it has to say:

§81. Les Japonais ont un grand nombre d’adverbes dont ils se servent non-seulement pour exprimer les modifications d’une action, mais qui indiquent encore le son, le bruit, la position de la chose. (…) On forme encore un grand nombre d’adverbes par la répétition du même mot, pour exprimer la manière dont se fait une chose, ou le son de cette chose : comme farafara, bruit de la pluie ou des larmes qui tombent. (p. 87)

[my translation:] §81. The Japanese have a great number of adverbs which serve not only to express the manner of an event, but which also indicate the sound, the noise, the position of the thing. (…) A great number of these adverbs are formed by repetition of the same word, to express the manner in which a thing is done, or the sound of the thing : like farafara, ‘sound of rain or of falling tears’

Here, we actually have a somewhat broader conception of the class — these adverbs are not mere imitations of sounds, they also express positions and manners. Moreover, we have a first morphological observation: many of them are reduplicated. Not all, mind you; most ideophone inventories known today do include a great deal of reduplicated words, but there are also plenty of morphologically simple roots. Incidentally, we’ve seen examples of both types before in the artful renditions of Kisi ideophones by Joanna Taylor: bákàlà-bákàlàthe sound of big, fat raindrops‘ and bíàààrain softly falling‘.


  1. Collado, Diego. 1632. Ars grammaticae Iaponicae linguae. [Project Gutenberg e-text]
  2. Spear, Richard L. 1975. Diego Collado’s grammar of the Japanese language. Center for East Asian Studies, University of Kansas.
  3. Rodrigues, João. 1825[1604]. Élémens de la grammaire japonaise [abridged from Arte da lingoa de Iapam] tr. et collationnés par C. Landresse. [With]. Trans. C. Landresse.


  1. I’m pretty sure there will be Japanese sources that go back even further, and offer more detail. Let me know if you have any![]

2 thoughts on Aduerbia sonus: Ideophones in two 17th century grammars of Japanese”

  1. I’d have guessed that the {x} in the example va va to xite stood for the hushing sibiliant, following from 17th century Portuguese-Spanish convention. So {shitte} in modern Romaji orthography rather than {itte}.

    Not that that really has any bearing on your point! :)

  2. Thanks! These days I’m posting just by way of taking a break from writing up something much bigger, so I didn’t have the time to check 17th century Iberian spelling conventions (though I suspected they would offer valuable clues) — much appreciated!

Leave a Reply

Your email address will not be published. Required fields are marked *