Beijing Sounds — 北京的声儿

A different way of thinking about how to put tones together

Let’s say you’re living in China’s Minnesota, the frigid dōngběi (东北 = northeast). You’ve come to Beijing for a bit of R&R and have met up with syz and his audio recorder. The idea is to talk linguistics and maybe record a random conversation or two along the way — fodder for the blog.

Suddenly after taxiing around Beijing for a bit, innocently explaining to the sījī (司机 = driver) where you’re from, syz interrupts your Mandarin train of thought with, “Hey, could I use some of this for the next episode of Zhonglish?”

What do you say? Was “Zhonglish” part of the bargain? Do you let the insult slide so as to achieve ever-lasting notoriety on the world’s most widely read (N=8) blog about the minutia of běijīnghuà?

Of course you do! Because syz has the most sympathetic ear you’ll ever find. His own thick tongue and disturbing (to him) inability to memorize hànzì make him eminently sympathetic to the difficulties of Mandarin. Anyway, calling your speech “Zhonglish” is no insult — it says so right here in the definition.

Thus in this episode I am pleased to present our first real-life VOLUNTEER for the role of Zhonglish guinea pig. This is important to note for a couple of reasons. First, it allows me to acknowledge feeling slightly guilty about the first issue of Zhonglish, in which innocent posters from Youtube were subjected, without anesthetic, to the knife in the Beijing Sounds studio. If one of them ever contacts me, I’m sure we can arrange the usual compensation package.

Second and more importantly, we now have a volunteering precedent. This allows me to revisit the challenge I’ve issued to a couple of our better-known Beijing-based blogger friends: send in your sound samples! In Brendan O’Kane’s case, the impetus was the downright hilarious recounting of the sījī who knew he was part Chinese (he’s not). For Chris Waugh, it was a tidbit about how he coulda shoulda recorded a phone conversation he overheard, where “the 马上 sounded more like mǎhàr than mǎshàng”.

The point is: should you ever be in a position to contemplate dissection of your own Zhonglish on Beijiing Sounds, fear not! It’s good for you. It’s cleansing — like a purge of the mental plumbing. Instead of the earlier “surgery” metaphor perhaps we should view it as… uh, well, use your imagination.

First tone = just aim high?

Despite the long intro, this entry is quite limited in scope: just some thoughts on how to approach tones when putting them together in a phrase. You certainly don’t need me to tell you that this is difficult. The teaching of tones is often overwrought (see item 1 in this entry) or poorly thought out — not based on real life. Tones are taught even less well in combination, and a combination with a third tone is especially difficult, it seems to wreak havoc on neighboring tones.

However, there are some good resources out there that cover much more about tones than I could possibly address here — and do it much more thoroughly. The ones I’m most familiar with are at John Pasden’s Sinosplice, which has both a blog and a great set of language resources. Laowai Chinese also has some good entries. These are good places to get started. Sinosplice also offers the amusing and accurate 7 stages of tone learning as a way to assess your progress.

In the taxi conversation in question, the recording just happened to include both our Zhonglish-speaking hero and the Beijinger taxi driver pronouncing hā’ěrbīn (哈尔滨, i.e. the northeastern city) one right after the other. The difference is quite striking. Here’s the snippet of conversation, but the words are also isolated in sound files below:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

The issue is clearly tones, but as I listened, it seemed like a visual would be useful. On my own I have occasionally tried doing just that — visually mapping the tones in recordings — using a very cool piece of open source software called Praat. Trouble is, the quality is usually too bad to see anything interesting. This time, though, it works. As you’ll see in the charts below, the pitch contours map very nicely to what you hear in the recording. (By the way, if you want to try this at home, John Pasden also offers a nice little tutorial.)

“Harbin” by Beijinger

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

“Harbin” by Laowai

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

What’s going on?

Two things: first, the facts of the situation; second, the psychology — why it’s happening.

One fact is that our intrepid informant misses the 3rd tone on the second syllable. I’m gonna argue that that is not a big deal. “Hā’ěrbīn” is a bit of an odd duck anyway. The conventional orthography of “Harbin” looks like only two syllables and the ěr kind of gets buried in the middle even by native speakers. But the other fact — and one that I DO think contributes to the Zhonglish effect — is the fact that his third syllable, the “bīn”, is low. It’s flat, as it should be, but it’s low instead of high like it should be, and that’s what sticks out. Contrast it with the taxi driver’s high-and-flat final tone — the sound as well as the visual are quite different.

I’m going to take a guess at the psychology as well, based simply on my own travails with the seemingly simple first tone. The trick seems to be to always put the first tone at the top of the range relative to nearby tones. A lot of times while making an effort just to keep the tone flat, I end up botching the overall delivery because I let some other tone go above it — even though I know all the textbooks tell me it’s a “high, flat” tone. This is a particular problem for me with the 1-2 combination with the 1-4 combination [thanks, Randy, for the correction -- this is what I meant]; I often want to start the second syllable above the first flat tone, thus making it easier to descend. It just doesn’t work.

Conclusion: my new favorite focal point is tone 1. Keep it high, relative to its neighbors, and make sure it doesn’t trail off, especially at the end of a sentence.

[Update 4/14/08 --

Here's a chart of the tones, copied shamelessly from Wikipedia. This is what Brendan refers to in the comments below when he says:

I tend to underemphasize it so that it ends up going from about 4 to 2 rather than 5 to 1

Probably everyone else knew what he meant but I didn't. It makes sense if you look at the right side of the chart and don't get it confused with the tone number. Thanks to Randy Alexander of Jilin City for the education.]

[update 4/16/08

Here’s another sample of the taxi driver’s speech, since Sima was asking (see comments) if we think he’s a real Beijinger. I’m gonna say: yes.

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.