Check out the essay, "Words as Feelings" on Aeon for some interesting stuff on ideophones.

Can you provide a link, please? I don't even know what an "ideophone" is.

I've not been able to paste links from Aeon for some reason.

https://aeon.co/essays/in-the-...he-word-was-embodied

Don't thank me, just send money.

Incidentally I read the article and I don't buy it. Without knowing exactly how the research was conducted it's impossible to assess its validity but other explanations for people guessing meanings more than chance dictates occurred to me immediately. One is the accidental similarity to words in their own language. For example "ribuy-tibuy" from Mundari meaning (so it claims, I'm also sceptical of such specific meanings) - "the sound or sight of buttocks rubbing together" wouldn't be a million miles from "rubby-tubby" and if they were asked "Do you think it means a) the sound or sight of buttocks rubbing together or b) coconut curry" I suspect that the similarity of sound would be an influencing factor.

Also, there is, in statistics, something called "the sharpshooter fallacy" which is to do with how data is collected. It's usually described like this. A gunman fires a large number of shots randomly at the side of a barn. Many hit, many don't. When he has finished he goes to the barn and finds where the shots have, purely coincidentally, formed a cluster, draws a target around it and says "Hey, look how often I hit the target."

Given that the article specifically says "Studies show that participants lacking any prior knowledge of Japanese, for example, often guess the meanings of the above words better than chance alone would allow", I wonder how many words in total were in the study. If it's a large number I would expect clustering - both of words that were guessed more frequently than "expected" and words guessed less frequently. Selecting the examples carefully enough could illustrate the point either way.

Another area where this could easily go wrong is this. Were the fake definitions supplied by native speakers of the original language or by native speakers of the study participants language? For example I could scour another language for words which sound a bit like their English equivalent and then create words that are not even close as the other choices. Like this...

The German for constipation is Verstopfung. It has "stop" in the middle of it so I could present the question either as

Does "Verstopfung mean
a) constipation or b) no entry? (either of which might seem plausible)

or

Does "Verstopfung mean
a) clarinet or b) constipation? (in which case I think the inclusion of "stop" in the middle would lead people towards the correct guess)

In short, I simply don't buy it.

This message has been edited. Last edited by: BobHale,

For anyone interested here is a great example of how the sharpshooter fallacy works. I can certainly confirm that there is no cheating going on in the video. (Well almost. I did spot one very tiny cheat which I can explain if anyone doesn't see it.)

https://www.youtube.com/watch?v=XzYLHOX50Bc

I give up - what?

OK. You will kick yourself because it is so obvious that everyone misses it.

The way it is done is this.

Toss the coin as many times as you need for there to be ten heads in a row. Maybe you will need to toss it thousands of times but chance dictates that sooner or later you will get ten in a row. Whenever you get a tail introduce the next toss with. "OK, here's the first toss." If that's a head introduce the next with "Here's the second toss." And so on. If you get ten in a row you have done it. Whenever you hit a tail before you have ten heads start again from "Here's the first toss".
Once you have ten edit the video so that it only shows that sequence of ten tosses.

And the very slight cheat is at the beginning. where he uses a slightly different introduction ("Ten heads in a row, watch") and then the camera does cut away so that introduction only needed to be filmed once.

This message has been edited. Last edited by: BobHale,

Aw - gee. I missed it.