"If you go after Mandarin, choose the somewhat uncommon GR over pinyin romanization if at all possible. It’s harder to learn at first, but I’ve never met a pinyin learner with tones even half as accurate as a decent GR user. Long story short, this is because tones are indicated by spelling in GR, not by diacritical marks above the syllables." http://fourhourworkweek.com/2007/11/07/how-to-learn-but-not-master-any-language-in-1-hour-plus-a-favor/

"Chao claimed that, because GR embeds the tone of each syllable in its spelling,[5] it may help students to master Chinese tones. One study however, found the opposite to be true in a study of GR." https://en.m.wikipedia.org/wiki/Gwoyeu_Romatzyh

  • 1
    Wow, never looked at GR before. GR looks really complicated. Seems like it would screw up pronunciation rather than help.
    – Mou某
    Commented Jan 22, 2016 at 6:04
  • 3
    IMHO the comment on the blog about Pinyin and GR is plain wrong and misleading, especially those who have no idea about it. The tonemarks are your friends! They show how the tone behaves over that syllable, level tone is marked by a flat line, a rising tone by a rising (acute) accent, etc. Nothing looks more natural than this. The GR system was conceived in 1925, long before Pinyin was around. Chao's claim was right at the time, because GR was better than any other system that transcribed Chinese with no tones or a more awkward system. But once Pinyin was introduced, it was superior to GR.
    – imrek
    Commented Jan 22, 2016 at 14:43
  • 1
    Another reason not to use 'Gwoyeu Romatzyh' is that nobody uses it for learning Chinese as of now. Dictionaries use Pinyin (and/or Bopomofo) so you need to be familiar with Pinyin anyway and if you are conversant in the Pinyin system why would you learn something else just to confuse yourself? You could learn both Bopomofo and Pinyin because they are distinct enough not to confuse them, and Bopomofo is still used in Taiwan as well as in the Xinhua dictionaries, but who uses this relict from 1925? Need to slap the author of that post with a stinky fish in the face for giving such 'advices'.
    – imrek
    Commented Jan 22, 2016 at 14:51
  • That's my sense too, but the preference for GR is shared by a scholar I respect, who says, "It is much better than Pinyin for the learner. Here's an interesting piece by him: eall.hawaii.edu/chn/chn451/03-Luomazi/GR.html
    – zadrozny
    Commented Jan 22, 2016 at 15:00
  • One more comment: you need to learn the tone of words by the ear, not by the eye. It does not matter if you can't recall if the tonemark goes up or down, as long as you can pronounce the word correctly, because then you also know which tone it is. The whole purpose of learning Chinese is to be able to speak it and that's just done by your ear in the most effective way. So I cannot really understand this quarrel about the visual debate over tonemarks and 'embedded' spelling. Pinyin is there, is used by almost everyone, so why resort to something that isn't used by anyone today.
    – imrek
    Commented Jan 22, 2016 at 15:10

5 Answers 5


In general, I think too much importance is ascribed to which transcription system is used. Students' ability to distinguish, pronounce and remember tones depends on many factors and the way the tones are written is just a very small part of it. Even if there was a difference between systems, it would probably drown in other factors.

The question of tonal spelling (i.e. including the tonal information in the spelling of the word rather than using diacritics) has actually been researched experimentally. Scott McGinnis (1997)[1] compared two classes taught with either Hanyu Pinyin or Gwoyeu Romatzyh. They were tested on reading tasks and the accuracy of their tonal production was assessed by native speakers. The study found that the Gwoyeu Romatzyh group actually performed worse than the Hanyu Pinyin group.

Personally, I think the problem is that tones are often treated as something extra, even optional. Not by all, but by far too many. Some teachers don't grade an answer as wrong if the tone is wrong (or left out) but the initial and final are correct, or they don't deduct as many points for tonal errors as they do for spelling errors. Students often ask "do we have to remember the tones too?" and at one point an intermediate student asked me if I was serious when I said that he had to learn all the tones (his tones were terrible and it turned out to be because he had just ignored them).

So, changing transcription system to Gwoyeu Romatzyh will not help. The little evidence we have suggests that it won't and there's nothing else that suggests that it will. Yes, it's good to treat tone as an integral part of a word, but you can achieve that by consistent and persistent teaching or learning methods instead. Therefore, using Pinyin is the best option since it's by far the most well-spread and accessible system.

Going off on a tangent, I do think that it's worthwhile for students to learn more than one transcription system after they have already learnt their first. I think that this is good because it breaks the strong grip of orthography. Seeing the same sound written with different letters or symbols may lead to new insights into that sound. I have no research to support this, but it was like that for me and I know many others who have had similar experiences when learning a second transcription system, regardless of which one it is.

[1]: McGinnis, S. (1997). Tonal spelling versus diacritics for teaching pronunciation of Mandarin Chinese. The Modern Language Journal, 81(2), 228-236.

  • You are entirely right about language teaching. The transcription method is not as important as keeping a good focus on pronunciation. But for other purposes, notably for publishing English (or other) language books about the Chinese language, pinyin is the right choice -- and it needs to be actual pinyin, with tone marks. Far too much toneless pinyin is used in scholarly works. Commented Jan 23, 2016 at 15:51
  • 1
    This is what I meant. I said "changing transcription system to Gwoyeu Romantzyh will not help", meaning that I think sticking with Pinyin is the preferred option. I have clarified my answer to remove any uncertainty about this.
    – Olle Linge
    Commented Jan 23, 2016 at 15:54
  • Instead of worrying about the transcription method, it'd be better if the government itself started to worry about simplifying (even more) some characters and adding tone clues in them... Chinese is way too much foreigner unfriendly. Commented Jan 27, 2016 at 12:30
  • That might help, but has there actually been any attempt by any government to simplify their own language to make it easier for foreigners to learn? Even if someone tried, I have a very hard time believing that it would work.
    – Olle Linge
    Commented Jan 27, 2016 at 15:05

Chao was a genius and a very funny guy. But pinyin draws on his experience with GR and a lot of other efforts. Pinyin is a better system. And of course it is far more widely used.

Tone marks seem weird to a lot of English speakers -- though of course they are standard in French too -- and used to be a big nuisance to type, on typewriters. But today they are easy on computers and people should learn to use them.


"I’ve never met a pinyin learner with tones even half as accurate as a decent GR user."

My only guess is that Tim Ferris has not met many Mandarin learners, then. Ask any foreigner with a near native leevel, one of those you see on TV or whatever, which system they learnt: chances are the vast majority will have studied pinyin.

Not to say that GR is bad. It's probably great and perhaps people who use it get good results with their tones: there will still be tons more people who have learnt pinyin and reached an equal level. In any case, they'll have to use pinyin to access the overwhelming majority of dictionaries (the remaining will anyway be in bopomofo, still not in GR).


I am a Gwoyeu Romatzyh supporter. I mean, who WOULD expect an English speaker to guess that 'x', 'q', 'zh', 'c', and 'z' to be pronounced the way that they are in Mandarin. GR romanizes based on pronunciation and the tones are easier to learn than at first glance. The tones on Pinyin may be left off as an afterthought and words like Běijīng and Bèijǐng could be confused for eachother. The GR versions are clearly different: Beeijing and Beyjiing.


Or maybe look into Bopomofo (aka zhuyin) which is taught in taiwan (so you can find plenty of children's books, dictionaries and computer/smartphone input methods that use it) . it's designed specifically for mandarin pronunciation and lots of people on this forum have said that it helped them with their speaking.

  • Thanks. I'm aware of Bopomofo (was in Taiwan for three months). It's interesting and probably a good supplement, but its lack of use outside TW (or for that matter on Google Translate) is a limitation.
    – zadrozny
    Commented Jan 22, 2016 at 14:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.