East Asia Student

Random Stuff Related to East Asia


Gwoyeu Romatzyh - better than pinyin?

gwoyeu / 國語

Gwoyeu Romatzyh. Have a guess what language that is. When I first saw it I guessed Mongolian, or maybe Korean for the first word. It’s actually Mandarin Chinese. More specifically, it’s the name of a romanisation system for Mandarin. A friend who also studies Chinese told me that when he started out, he used this ‘Gwoyeu Romatzyh’ system instead of pinyin. He reckons, and he’s got Tim Ferriss on his side for this, that using GR is better for learners, particularly when it comes to mastering tones.

What’s wrong with other systems?

There are countless systems for writing Chinese with the Roman alphabet. A big problem for any romanisation system is how to handle tones. They’re completely essential to the language, but Roman letters weren’t designed to encode information beyond vowels and consonants. The Wade-Giles system a number in each syllable, e.g. guo23, and pinyin uses tone-marks, e.g. guóyǔ.

The pinyin tone-marks system looks elegant and compact, and is ultra-consistent. However, a lot of devices display tone marks incorrectly, or not at all. Even worse, there’s a temptation to not even use the tone-marks in the first place, as they seem like a bit of an afterthought rather than something essential.

In an ideal world everyone would write Běijīng and Máo Zédōng and the tones would be preserved. But it’s so easy to just write Beijing and Mao Zedong (and argue that those are in fact the English spellings of those names), that a lot of Chinese words end up getting written without tones.

With the Wade-Giles number system, the tones are at least a bit more obvious (Pei3ching1, Mao2 Tse2tung1). Unfortunately, they’re a bit too obvious and look clumsy, suitable only for textbooks and detailed linguistic explanations. So again, they don’t get used as they were intended.

What does Gwoyeu Romatzyh do differently?

Gwoyeu Romatzyh’s solution to this problem is to actually incorporate the tone into the spelling of each syllable. Different letters for different tones. So the words above become Beirjing and Mau Tzerdong (yes, I found ‘Tzerdong’ pretty funny as well). All of the information for each syllable is contained there: consonants, vowels and the tone. Pinyin.info has a more detailed comparison table.

There are some immediate advantages to this idea. The biggest is that you have to include the tone when writing. There’s no way to spell things without the tone. So you as a learner you wouldn’t have the frustration of seeing things written down but not knowing how to pronounce them properly (this happens all the time with Chinese people’s names written in pinyin).

The second advantage, and the one that’s most interesting here, is that Gwoyeu Romatzyh makes it easier to remember the tone for words you’ve learnt. Suddenly, remembering the tones for Chinese words is no harder than remembering the spelling of words in European languages. This really does work, according to the friend who told me about the system.

So is it better for learners?

Tim Ferriss, master of exaggerated headlines, recommends that people learning Mandarin use Gwoyeu Romatzyh and not pinyin:

"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." _[How to Learn (But Not Master) Any Language in 1 Hour](http://www.fourhourworkweek.com/blog/2007/11/07/how-to-learn-but-not-master-any-language-in-1-hour-plus-a-favor/ "How to Learn (But Not Master) Any Language in 1 Hour (Plus: A Favor) - Tim Ferriss") - Tim Ferriss_

I would say that’s just because Tim Ferriss hasn’t met a lot of pinyin learners (also notice the no true Scotsman trickery with the addition of ‘decent’). The problem with this view is that it assumes the main way you learn tones is by remembering the spelling. This may be true in the very early stages, and if that’s what Ferriss is referring to then he may be right.

But if you’re learning a language for any real length of time, you just remember words by their sound. There are a lot of illiterate people in the world, but they still remember how to speak their language as well as anyone else. Once you’ve actually acquired a Chinese word, you do remember its tone because that’s just how the word sounds.

Romanisation systems are really just a crutch for people learning Mandarin. You use them almost constantly at first, then gradually less and less until the only time you need them is when you look up an unfamiliar character in a dictionary. By that point, it really doesn’t matter how the tone is indicated, so long as it’s there. You’ll remember it.

The Gwoyeu Romatzyh study, and its flaw

The potential benefit of Gwoyeu Romatzyh for learning tones was addressed specifically in a study at the University of Maryland in 1997 ( The Modern Language Journal Vol. 81, No. 2 (Summer, 1997), pp. 228-236 ). They got a group of non-native speakers who were studying Mandarin, some from America and some from Japan, and had one half study with pinyin and one half with Gwoyeu Romatzyh.

After a year of study, they had the students take a pronunciation test and graded their tonal accuracy. The pinyin learners actually did better than the GR group. This would’ve been an interesting result, except for the fact that they tested them using romanised Chinese. I’m not at all surprised that people reading out pinyin, which has clear marks for each tone, would have less trouble remembering the tones.

What they should have tested was tonal accuracy when reading out Chinese characters. That would actually test whether Gwoyeu Romatzyh helps you to remember tones correctly. As it is, we’re still in the dark about this.

Even if GR is better, don’t use it

I seriously doubt that Gwoyeu Romatzyh is better in the longterm, for the reasons given above. But even if it were, I would still recommend that people use pinyin. The friend of mine who started out with GR actually ended up switching to pinyin anyway, despite believing that GR is better for learning.

He did this because pinyin is now the dominant standard for romanising Mandarin Chinese. It’s pretty much everywhere, and if you’re trying to use some other system it just gets confusing. Standards are good, and it’s worth sticking to them once they’re established.

Pinyin is actually better anyway

In any case, I’m almost certain pinyin is better. It’s better because it’s consistent. It’s a standard with standards. Once you know how to pronounce ei in Mandarin, you can consistently get that sound right wherever you read it in pinyin. It won’t get mangled into something else, and mislead you into thinking its a different vowel sound.

Because pinyin is set up like this, you can isolate the sounds of Mandarin and learn them in the shortest possible time. You can see the patterns in the pinyin table and use that to your advantage.

Further reading