5

I refer to this map (also shown below), which depicts that Yue and Hakka variants of Chinese are geographically very close to each other.

If we take Hong Kong for example, how similar would the language be to the Hakka spoken in the neighbouring parts of Guangdong? (see red circle below)

enter image description here

  • 1
    There is a dialect xingning (興寧), which is a cross between Hakka and Cantonese. It has some words that sound Hakka and some words that sound Cantonese. – cup Feb 22 '18 at 15:00
6

If we take Guangzhou to represent Cantonese and Meixian to represent Hakka we can get an idea.

Part one:

Isolated-word-intelligibility test

IWIT

  • With Hakka as the speaker people from Guangzhou only understood 35% of isolated words.

  • With Guangzhou speakers Hakka listeners only understood 40% of isolated words.

Part two:

Sentence intelligibility test

SIT

  • Guangzhou speaker were only able to get Meixian listeners to understand 55% of spoken sentences.

  • Meixian speakers were only understood 12% of the time by Guangzhou listeners in spoken sentences.


The standard for mutual intelligibility works like this:

If two languages are 80% intelligible (or more) with each other, then they are dialects of one language rather than two separate languages.

These two certainly don't meet the requirements.

It seems that Hakka speakers can understand more Cantonese than vice versa.


If you'd like to know more about the intelligibility tests you can check out this PDF here. Where section 3.1. Isolated-word-intelligibility test goes over table 2 and section 3.2. Sentence intelligibility test goes over table 3.

  • This is exactly what I was looking for. As for your last sentence: "It seems that Hakka speakers can understand more Cantonese than vice versa.". Any ideas as to why this may be? I am having a hard time understanding why the numbers should not balance. The imbalance makes it hard to judge the similarity. – udidosa Feb 22 '18 at 5:34
  • @Urvil see the Wikipedia article on asymmetric intelligibility – droooze Feb 22 '18 at 6:48
  • 1
    @Urvil It does not have to be balanced - some dialects are more sloppy and less prevalent in movies and culture. Danish people can understand more Swedish than vice versa - Sweden is a larger country and produces a bit more TV-series and things like that, thus making Danes more familiar with Swedish. Also, the Danish language has incredibly unclear pronunciation. This is not subjective - studies cphpost.dk/life-in-denmark/… show that the Danish language is objectively harder for kids to learn. – Per Alexandersson Feb 22 '18 at 9:16
  • 1
    The standard for mutual intelligibility has a problem when A & B can understand each other 85%, B&C can understand each other 85%, but A & C are only at 65%. – 伟思礼 Feb 22 '18 at 14:13
  • 1
    @Urvil You could almost make a new question for that. You're right Chaozhou speaker - Meixian listener gets 15%; Meixian speaker - Chaozhou speaker gets 25% on isolated words. On sentences it drops dramatically for Chaozhou speakers - Meixian listener only gets 3% (say what?); Meixian speaker - Chaozhou speaker gets 20%. – user3306356 Feb 23 '18 at 0:48
3

Quoting from wikipedia:

Hakka is not mutually intelligible with Yue, Wu, Southern Min, Mandarin or other branches of Chinese, and itself contains a few mutually unintelligible varieties.

The same could really be said of any of the southern Chinese language groups. Lots of linguistic diversity in a small geographic area (in comparison to Mandarin languages/dialects, which cover a very large area and are often quite similar to one another).

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.