Why do some languages sound so fast?
posted by Sandeep
To unfamiliar ears, many foreign languages sound unintelligibly fast. Does this have to do with something inherent in the language itself, or is it just because we don’t know what the other speaker is saying?
A study that appeared in the journal Language is summarized in TIME today. It attempts to answer this question with the following experiment:
To investigate this puzzle, researchers from the Universite de Lyon recruited 59 male and female volunteers who were native speakers of one of seven common languages — English, French, German, Italian, Japanese, Mandarin and Spanish — and one not so common one: Vietnamese. They instructed them all to read 20 different texts, including the one about the housecat and the locked door, into a recorder. All of the volunteers read all 20 passages in their native languages.
Read the article in TIME to find out what the investigators found. Really fascinating stuff!
johnwcowan, 
Bathrobe, 
Stan 4:04 pm on September 8, 2011 Permalink |
The full paper is available here (PDF).
Welcome, by the way! I’m enjoying the blog.
The Diacritics 10:30 pm on September 8, 2011 Permalink |
Thanks for the link!
stuartnz 6:32 pm on September 10, 2011 Permalink |
Thanks for a very interesting read, I look forward to many more. Is it possible to get a link to the script map you use in your blog’s header?
The Diacritics 6:52 pm on September 10, 2011 Permalink |
Hi stuartnz, the map is a public-domain image available on Wikimedia Commons. Here’s the link: http://commons.wikimedia.org/wiki/File:WritingSystemsoftheWorld2.png. It’s missing a couple of minor scripts from Asia, but otherwise it’s a pretty great map. Enjoy!
stuartnz 6:56 pm on September 10, 2011 Permalink
Thanks!
stuartnz 6:59 pm on September 10, 2011 Permalink
Studying that map, I see it’s also missing scripts like Nastaliq. It seems that Pakistan, Afghanistan and Iran are all blank on the map.
The Diacritics 7:05 pm on September 10, 2011 Permalink
Good catch. It seems like they glossed over Nastaliq because it’s based on the Arabic script, but it should probably still be on there. I suggest contacting the creator of the map.
The version I sent you also had malformed Kannada and Telugu labels. Here’s the most recent version: http://en.wikipedia.org/wiki/File:WritingSystemsoftheWorld4.png
stuartnz 7:10 pm on September 10, 2011 Permalink
Thanks for the link to the revised version. I figured that Nastaliq had been left off for the reason you suggested, but will contact the creator to see. Since Nastaliq and naksh are modified derivatives of another modified derivative of Arabic, maybe all the Indic/Dravidian scripts should have been lumped in under Brahmi.
César 7:00 pm on September 10, 2011 Permalink |
A very interesting read, indeed. Thank you.
[After reading some of your (plural) posts, I am a subscriber now. Maybe I am the first one from Spain]
Bathrobe 7:46 pm on September 10, 2011 Permalink |
I had a lot of trouble with that test.
This one left me gaping: “further analyzed how much meaning was packed into each of those syllables”. How do you objectively analyse how much information is packed into a syllable? For instance, how do you calculate the amount of information packed into xǐshǒujiān vs o-tearai vs toilet or bathroom? The Chinese word has more information (wash-hand-room), the Japanese has less (wash-hand). English has less, sort of (bath-room). How do you calculate the information density of syllables like this?
The conclusion that “Spaniards really do sprint and Chinese really do stroll, but they will tell you the same story in the same span of time” just doesn’t make sense. Even in English there are speakers, possibly whole dialects, that speak slower than others. At the pragmatic level, Chinese might have “high density”, but that doesn’t tell you how people speak in everyday life. People don’t necessarily speak in just the number of syllables required to get the message across. Information-dense Chinese can take longer to get the same message across, depending on how much your speaker has the gift of the gab. Beijingers are notorious for their prolixity and they speak fast, but information density isn’t necessarily that high.
Sorry, the study raises more questions than it answers for me.
johnwcowan 11:23 pm on September 10, 2011 Permalink |
There is also the problem that all the texts are translated from English (except the English ones), and translations are not typically the same length as spontaneous texts — typically they are longer, as the translator must disambiguate things that can remain safely ambiguous (because only one meaning is relevant) in the original. Drawing on a large stock of translations in multiple directions would have been appropriate in order to develop a correction factor.