Friday, October 14, 2011

GraspChinese review: Pretty good way for Chinese beginners to get started is an online Chinese course for the total beginner. The first 10 lessons are free and for $14.99 (or $12.74 with the 15%-off promo code "GRASP15" that will work through the end of 2011) you can get a month of access to all 37 lessons, with discounts for longer subscriptions (GraspChinese provided me with a free six-month subscription for review purposes, so I had access to all 37 lessons). The short and succinct lessons cover the basics of Chinese vocab, grammar, and pronunciation, with a welcome focus on teaching the four tones. It's a pretty good place to start your Chinese efforts, especially given how quickly you should be able to go through the material, but wholly omitting Chinese characters is disappointing and needing to hear a non-native speaker pronounce Chinese words is not ideal.

GraspChinese's lessons are for the most part organized around doing a bunch of things that a traveller to China might need to do while there (ordering a coffee, shopping, etc.). Each lesson consists of a series of narrated slides, a listening comprehension exercise, dictation exercises for words and phrases, a list of vocabulary and key phrases from the lesson, and—when the vocabulary of the lesson lends itself to visual representation—an image selection exercise not unlike those found in RosettaStone or Livemocha.

The teaching method used in their lessons reminded me most of that of Michel Thomas, whose method introduced what seemed like a relatively small amount of simple vocab and phrases to quickly get the learner to say more and more useful and complicated things. GraspChinese's lessons largely follow that approach, with perhaps a bit more explanation than Michel Thomas. One of the things I liked about GraspChinese's method was that they often use multiple explanations explain the same piece of information, which leads to more ways for your brain to make the recall. For vocab, they'll tell you the meaning of the word and what the components of the word means literally, and they also sometimes explain patterns in word formation (e.g., 餐 cān ("meal") being paired with words for "early", "noon", and "late" to make 早餐 zǎocān ("breakfast"), 午餐 wǔcān ("lunch"), and 晚餐 wǎncān ("dinner"), respectively), provide tips to help remember (e.g., 吃 chī ("to eat") sounds like "chew"), and cover mistakes that an English-speaker might make (e.g., ordering 鸡 ("chicken") in a restaurant instead of 鸡肉 jīròu ("chicken meat") might make people think you want a live chicken). Grammar is given to you via sentences, with a literal, word-for-word translation of the sentence, a standard translation, and brief explanations of the grammar points (and, no, they thankfully don't cover gender).

One of the strengths on the lessons is their focus on pronunciation. They pause to explain points that'd be particularly confusing for English speakers (e.g., the pronunciation of c, zh, etc.), but more importantly they focus on the tones (something my own teachers neglected, which to date still affects my Chinese). They repeatedly note which words have which tones, pause to distinguish between places where tones are all that differentiate spoken words (e.g., 哪里 nǎlǐ ("where") and 那里 nàlǐ ("there")), and even note where English uses the tones. For instance, they note that the rising second tone is just like how English speakers raise the tone for "What?" and that the falling fourth tone is much like the falling tone in English commands like "Run!" Pointing out how we actually also use tones when speaking English (even if as part of intonation rather than to differentiate words) goes a long way in demystifying the tones for Chinese learners, and it's a point that often seems to be overlooked.

While GraspChinese's lessons do a good job on pronunciation, the lessons and the site overall fall flat in respect of Chinese characters. They focus on speaking, so I can give them a pass on not explicitly trying to teach the characters. However, the site makes no use of them whatsoever, instead relying solely on pinyin, which I think is a huge wasted opportunity. As GraspChinese explains it:

Pinyin is the method of writing Chinese in the letters of the English alphabet. It is the starting point for millions of Chinese schoolchildren every year as they start their formal education. Our approach to teaching Chinese is based on using pinyin too. The Chinese characters are complex to learn and are not essential to start learning and communicating in Chinese.
I'd debate both that Chinese characters are "complex" to learn (learning Chinese characters requires effort, digging a ditch requires effort, and neither are complex) and that they are inessential to even a beginner (nothing will quite cement things for you—even at the beginning—as seeing how the characters link up with the words), but without getting into those arguments any further, why not just include both characters and pinyin, even if there's no concerted effort to teach the characters? That way, users could still rely on the pinyin crutch even as they get a modicum exposure to the characters.

Just to provide a quick example of how simple this would be, take this original image with pinyin and one of their word-for-word literal translations:

Now take a wee bit of my own image-editing magic and, bam, Chinese character exposure, at no loss to the existing content:

To turn back to the lessons, they are wholly in the form of a dialogue between Leo (the American character who is the Chinese learner in the lessons, who at first I thought had adopted the Chinese surname 刘 Liú ("Liu") before I realized his name was "Leo"), Dani Wang (the founder of the site and the main teacher in the lessons), and other characters who drop in based on the situation called for by the lessons. The situations are contrived—although pretty representative of commons things you'd run into in China—and the dialogue somewhat corny, as is the norm for lessons like these, but I somewhat strangely found myself enjoying them (on the other hand, my wife, who passed by as I was listening, heard Leo speaking and described it as "painful", so your mileage may vary). The dialogues, in any case, do serve their purpose of creating a platform from which Chinese can be taught.

One aspect of the lessons that I'm not a fan of is repeatedly listening to presumably non-native Leo pronounce Chinese, a fault actually shared with Michel Thomas products. He's clearly got pretty good pronunciation, and I'd imagine that they did enough takes of each recording to make sure that was the case, but he still seems to sound foreign. Now if that's just acting and they in fact have a truly native-level bilingual person doing the recordings for Leo, that's great, but my suspicion is that it's a coached non-native speaker, and with limited exposure to the language you need to get the most bang for your buck, which would come from listening to a native speaker at all times, rather than the supposedly confidence-building exercise of hearing another learner "get it".

Any good language-learning scheme will work from the most frequent words to less frequent words, and GraspChinese generally seems to do that. However, there seems to be a number of outliers that would seem out of place if they hemmed more closely to frequency lists. For instance, slide 37 of lesson 8 introduces us to the word 培根 péigēn ("bacon"). I don't recall ever seeing bacon in China outside of hotels with foreigners, so this doesn't seem like the ideal choice for inclusion in lessons aimed at beginners. Similarly, lesson 11 brings you "project manager" and "marketing department"; although certainly more useful than bacon, I'd venture that you could come up with a dialogue with more commonly used words.

Any good language-learning scheme will also repeat previously learned material. Between lessons and exercises, GraspChinese does seem to make an effort to do this, although the approach seems to be ad hoc, as opposed to, say, the Pimsleur recordings that systematically repeat words with a repetition timing based on the forgetting curve.

As noted above, once you get through the lesson, there are a series of exercises relating to that lesson. The listening comprehension exercise presents you with a word or phrase in English and you have to select from four recordings in Chinese the one that matches with the English, and it seems that if two words or phrases are distinguished by tone only, those two tone variants will always be options, which will help to ensure that you can hear the tones properly. In the dictation exercises, you listen to a word or phrase in Chinese and then must type it out—again with tones. Image exercises consist of a word written in Chinese with matching audio and you must select which of four pictures matches up with the word.

The most annoying problem with the exercises is the dictation exercises' lack of tolerance for certain irrelevant errors. "Tolerance", when talking about flashcard or similar programs, is the ability of the software to ignore mistakes that don't actually indicate a lack of understanding, and there are two kinds of those that come into play on GraspChinese. The first is spacing. Actual Chinese has no spacing; Chinese characters are just strung together one after the other and context plays the role that spacing plays in English. However, when writing in pinyin, spaces are typically used as they are in English, although the rules in Chinese on where one word ends and the next begins—i.e., where spaces should go—can be fuzzy. In GraspChinese, if you don't master the arcane art of determining where the spacing goes, you'll get dinged for a wrong answer. For 你好 ("hello"), GraspChinese marks nǐhǎo as wrong and nǐ hǎo as right. For 飞机场 ("airport"), fēijīchǎng is wrong and fēijī chǎng is right. When negating a verb, a space is used with 不 ("don't") but not with 没 méi ("don't"). Spacing should be ignored completely, as it has no bearing on whether the learner knows the word.

The second is tones. GraspChinese has a very simple method to add tones; type the vowel, then press 1, 2, 3, or 4 to add the appropriate tone. That sounds great, but it requires you to pick the specific letter on which the tone appears, even for diphthongs (i.e., two vowels in the same syllable). Here again, there are arcane rules as to which of the two vowels in a diphthong gets the tone. In 飞 fēi ("to fly"), it's the first vowel, but in 所 suǒ ("place") it's the second. I have no idea why it'd be one or the other, but it seems silly to mark it as wrong when the correct tone is simply over the wrong vowel in a diphthong (as I did with the suǒ in 厕所 cèsuǒ ("toilet")). The way I'd like to see this resolved is for the tone to be moved to the appropriate place automatically if the user puts it on the wrong vowel in a diphthong.

Some thought was put in to making these exercises user friendly, such as keyboard shortcuts, but there were a few pain points where the process could be made even smoother. For instance, when you first switch to one of the dictation exercises that require you to type, your cursor isn't in the typing box, so rather than just starting to type you need to first click in the field, after which you can just type through the rest of the exercise. Similarly, when you end an exercise, you have to click on the next one to keep going; as GraspChinese recommends what the next step should be, I'd much prefer to simply press return to quickly move on to the next item. More significantly, there isn't any way to review what you got wrong after you have completed an exercise; at the end, it tells you your percentage, but you can't review your specific incorrect answers anywhere. (As an aside, as I was going through the lessons and exercises, it crossed my mind a few times that a full-screen mode would be nice, so I wouldn't have to look at all the other distracting crap floating around on my computer.)

GraspChinese attempts to direct you through their lessons with a recommendation for what to do next after each lesson or exercise. However, it ends up not being very useful because of the way it works. Basically, an exercise is not removed from the queue until you've gotten 100%. Even if you got 95% (and, frustratingly, even if that one wrong answer was because of a mistaken space), it will always direct you to the first exercise that you haven't gotten 100% in yet. The result is that, if you follow the suggestions, you will always stay in the same exercise until you get 100% on it. That sounds like a recipe for either frustration or simply ignoring the suggestions (which I eventually came to do as I went through the lessons).

GraspChinese is a relatively new product (the full 37-couse set was only released on July 9, 2011), so unsurprisingly there are still a number of rough edges at the time of my review. When using Safari, slide 7 of lesson 6 and the other slides using the same image have what appears to be the negative of the intended image for the character displayed, which is jarringly pink (and kind of amusing in just how blatantly out of place it is), although the same image appears properly in Firefox and Chrome. Slide 39 in lesson 12 mistakenly had a fourth tone where a neutral tone should be, although this was quickly fixed when I pointed it out to GraspChinese (which is much more than I can say about errors pointed out to Livemocha). You can't simply click the back arrow to rehear the last slide in a lesson; doing so will take you to the first card, and then you have to click (through dozens of slides) all the way to the end to hear it again. You can click on where the last card appears in the progress bar to go back to the last card, but that solution wasn't obvious to me at first and doesn't change the counterintuitive behavior of the back arrow.

I also noticed a couple of places where Dani's pronunciation of English is mistaken or unclear. In slide 29 of lesson 8, she pronounces "buffet" as "boo-fate", even while Leo pronounces it correctly in slide 57 of the same lesson. In slide 69 of that lesson, which happens to be the lesson in which Leo learns how to say the word for the decidedly non-kosher "bacon", she seems to say "You know all the kosher words now". At first I thought it was some kind of joke, but after listening to it a few more times it seems that the intended word was not "kosher" but "crucial". With these kinds of issues in the English parts of the dialogue, it seems like an English proof listener could be a good investment, especially to make this more readily accessible to non-native-English speakers, who might not be able to figure things out so easily with a few extra replays.

The last component of each lesson is the "phrasebook", which is just a list of the vocab and the key phrases from the lesson. This comes with clickable audio, but sadly comes with no way to review the items in the list. I'd love it if there was a way to download this to Anki—complete with the audio—but as is it's just a static list.

One of my favorite features about the site overall is that it's quick. The lessons and the exercises are short and it doesn't take much time to watch a lesson and run through the accompanying exercises; you could probably tackle a set on a coffee break. That said, it feels like you're covering a lot of ground, and covering a lot of ground in a short time is always the goal.

You may recall that I'm not a big fan of not being able to easily access pricing information, and GraspChinese's pricing information is not very easy to find. You need to click on one of the lessons that are not free, and then you'll see a "Subscribe" link that'll take you to the pricing information; searching for "pricing" or "subscription" on the other pages will get you no hits at all. In any case, the first 10 lessons are free, while it's $14.99/month for a subscription to all 37 lessons. Given how quickly you should be able to move through these lessons, I'm not sure what the point of paying for more than one month would be, but a six-month subscription is $29.99—a huge discount off the monthly price. There is also a 12-month "executive" price, which includes 24 one-to-one audio sessions with one of GraspChinese's "professional tutors" for $299.99/year. Few details are provided on this tutoring, but assuming each class is an hour long, you're paying at least $10 per hour of class, and I'll bet you could get a better price for similarly qualified teachers on sites like eduFire. Until the end of this year, you'll also be able to use a promo code, "GRASP15", which gives you 15% off, reducing the above prices to $12.74, $24.49 and $254.99, respectively.

All in all, I'd say that GraspChinese is a pretty good way for a new Chinese learner to get his or her feet wet. It's not going to get you to fluency (and it doesn't pretend to), but it'll cover some of the basics for you in a way that feels quick and light and, considering that you should be able to get all of its content for under $15, the price point doesn't seem unreasonable either.


  1. I love how thorough this review is & hope they take some notes!

  2. When I first started studying Chinese, I had a grammar book that used only pinyin. And while I understood that they weren't going to teach the characters, I always thought it was silly that they would even let me see them, along with the pinyin, as if it were just too much for my noobish mind. As noted above, I think GraspChinese is fine in putting off the proactive teaching of Chinese characters until later, but there's no reason not to just let the learners see them. It doesn't take anything away from the information you're already trying to teach, and it'd help learners get used to looking at characters. And perhaps they might even pick up a few characters along the way.

    Given that pinyin is not ultimately used in Chinese, being a stickler for pinyin orthography rules seems kind of pointless; having no clear idea of what the rules are after years of Chinese doesn't affect my ability to communicate at all (unlike, e.g., issues with tones). So I would say that it's good that you follow the rules strictly in your presentation, but there's no point in dinging users for not knowing them when doing so has no affect on their ultimate ability to communicate.

    If you insist about being strict about pinyin orthography, then you should at least explain the rules as part of the course. I don't think it's reasonable expect that users who are looking to learn Chinese quickly are going to take a detour through an esoteric Wikipedia page to figure out how to properly use spaces and accents with pinyin.

    Given that you will always have a limited amount of exposure to a language, that exposure should always be of as high a quality as possible, and you lose some quality when listening to a non-native speaker. A native speaker might be demoralizing if he or she is clearly a native speaker, but if you found someone whose English and Chinese were native but could act just as Leo does—pronouncing out words slowly and deliberately, asking how things work, etc.—I doubt learners would even realize he or she was a native speaker, much less be demoralized by it. And, by doing so, learners would then get the benefit of 100% of their exposure to spoken Chinese being from native speakers—even though one of those speakers is acting as if he or she were not.

    On the user of the button, it only takes you back to the first card if you've already finished the set. Otherwise, it works as you described. Thus, the only slide you can't click back to is the last card in the set, as noted above. I tried that out on Safari, Firefox and Chrome on a Mac, and it was the same in all three, which does not seem like browser-dependent behavior.