The mysterious saw

The Japanese word for saw is nokogiri. The -giri is transparently a nominalized and voiced form of the root /kir/, "cut". So what is the noko?

My first guess was that it was somehow related to /ki/, "tree/wood", which regularly appears as /ko/ in compound words like kogarashi*, as discussed here. There's even a word noki, meaning a tree standing in a field (no), and a homophone (homophonic status in OJ times uncertain) that means "eave". Any of these things, a saw can cut. The only remaining question is which it was originally. Case closed.

But no! A trip to the dictionary reveals that the word was originally nohogiri. Ki, wood, is irrelevant. What, then, is a noho?

Sadly, no-one knows for sure. Many have proposed some relation to the /nob/ in modern words like nobiru ("stretch") and nobasu ("extend [something]"), etc., but this theory is undermined by the fact that /nob/ isn't found as /noh/ anywhere else. To patch this leak, some insert a /ha/ ("edge", "tooth") in between the /nob/ and /kiri/ and implicitly invoke a poorly-understood Blender Principle to get the final word -- but you're asking for an awful lot of sound change at that point, finalized and traceless before writing began.

All this etymological opacity, plus the fact that the referent is non-obvious technology, obviously raises the possibility that the word's origins lie outside of the Yamatosphere altogether. If this were the case, the variant noko might be not an abbreviation but rather a descendent of the original word itself, with nokogiri just meaning "noko (or noho, nopo, etc.) cutter".

In from the western ports they rode, reins clutched in one fist and carpentry tools held high. O fierce ones, I ask you then: What is best in life? To build wooden dwellings, to plane them smooth -- to carpet the land in sawdust and hear the lamentation of the native construction industry. That is what is best!

* Kogarashi: "Tree (/ki/) witherer (/karas/)" → "Cold autumn/winter wind that blows leaves off trees". Also a woman's word for "pestle", maybe via phonemic play on the standard word surikogi, although Ōno Susumu claims (offering no sources) that it's because a pestle is smooth and bald like a post-kogarashi tree. Pfft. (Back)

Popularity factor: 5


For obscure reasons I won't get into, I had to do some research into the origins of woodcutting tools in Japan. There was some controversy about the earliest date that flat wood boards were produced. It appears that flat boards were originally planed with chisels or adzes rather than sawn. So perhaps the noko- is a "stretched or extended cutter," in that it is a long saw rather than a narrow chisel or adze. But this is purely speculation on my part.

Paul D.:

I don't suppose anyone's compared the words for tree, wood, and branch in classical Chinese dialects and Old Korean for clues to the identity of noho?


Charles -
Or in that it makes a long cut?


Yeah, the reason why it could be related to /nob/ meaningwise is pretty clear... the difficulty is how to get there from here in terms of sound change. (Another idea that struck me was that it could be the /oho/ meaning "big", but then explaining the /n/ is problematic.)

Paul D: Not to my knowledge (thesis!). I do know that the Ainu for "saw" is /noko/, but I presume that they got it from the Japanese-speakers (if there was any evidence otherwise you'd think that someone, somewhere would have recorded it).

language hat:

In the absence of John Emerson, I will maintain that this is irrefutable evidence of the Dravidian origin of Japanese.

Comment season is closed.