After months of exile in Minneapolis, Siberia, the nominal proprietor of Beijing Sounds now has concrete plans to return to the main studio location in beautiful suburban Shàngdì in the northwest outskirts of the capital city. The mood surrounding this semi-permanent move, to take place around the beginning of April, is in great part exuberant: reunion with PBS and Mrs. BJS, many a good meal at YU, along with a little more coursework… But inevitably a bit of nostalgia will slip in: a job and coworkers so good that they are difficult to leave behind, a few friends who are far better than any socially stunted recluse deserves… And then there is two-and-a-half year old Cici, who has developed a fondness for this shūshu (叔叔 = lit. “uncle”, a form of address) that’s utterly endearing.

Lucky, then, that the studio recorder was turned on last night to catch her at a New Year’s party and give shūshu a departing gift of inspiration. Straight from the mouth of the babe…*

[Go to website or bottom of this post to listen to audio]
Father:
中文怎么讲？”Where’s mommy?”
Zhōngwén zěnme jiǎng “Where’s mommy?”
How do you say “where’s mommy” in Chinese?

Cici:
“Where’s mommy?” [in English, obviously]

Father:
中文怎么说？
Zhōngwén zěnme shuō?
How do you say it in Chinese?

Cici:
哪里是妈妈呀？
Nǎlǐ shì māmā ya?
“Mommy is where?” [using the wrong verb -- shì, the equivalent of English "is" instead of zài -- as well as borrowing the English word order]

[laughter]

Father:
“妈妈在哪里”!
“Māma zài nǎlǐ”!
“Where’s Mommy?” [correcting grammar]

Cici:
妈妈在哪里？
Māmā zài nǎlǐ?
Where’s mommy? [repeated with correct grammar]

Father:
对！ 不是“哪里是妈妈”
Duì! Bùshì “Nǎlǐ shì māmā”
Right! It’s not “Mommy is where?”

Fantastic. Even Kingsoft 2002 couldn’t get you a better word-for-word translation.

So where does a two year old get this? You might guess her Mandarin is simply weak in general, or at least perhaps weaker than her English, and that’s why the English word order dominates.

Hardly the case. She does well with English, especially comprehension, but her Mandarin is more fluent. Her parents typically speak pǔtōnghuà** (普通话 = standard Mandarin) together, and her mother stays at home with her full time — no daycare.

Regarding English, exposure seems limited to a few playdates with English-speaking friends and watching some TV.

Given this background, you might think the English would be playing sorry second fiddle. Yet there it is, wreaking havoc with the vocabulary and word order of the simplest Mandarin sentence. If this is how hard a two-year-old has it, maybe the rest of us Zhonglish*** speakers should just pack up our Plecos and call it quits.

—————-

* In keeping with the money-back guarantee in the sidebar, “All sounds are from real events and situations unless otherwise noted”, in-house counsel has strongly recommended noting a potential gray area. Cici, who is both mischievous and wise in nearly lethal doses, has made this grammar error before, presumably in good faith. In this instance, however, she is in a sense being asked to repeat it. As she is very in tune with all activities which cause her parents consternation and in fact relishes discovering such activities so that she can repeat them at will, it is within the realm of possibility that she is, in the vernacular, “hamming it up” for the crowd. In other words, knowing how concerned her parents are with this botched phrase, she might now be much more likely to repeat it even though she might know, on some level, that it’s wrong.

** Of possible linguistic interest is the fact that when she was one year old she spent several months in China near Shanghai and may have been exposed to more Wu than Mandarin. But I can’t see that it explains anything.

*** [update] Links to all Zhonglish articles

All roads lead to -ngr

The email discussion on -ngr started soon after the claim was made, on these pages, that -ngr was a sublime consonant. In a spasm of curiosity, a member of the Beijing Sounds staff had written to phoneticist John Wells to try to get a better understanding of just what that érhuàyīn was doing — both in the nasal-oral cavities and in the syllable. John Wells CCd several other eminent linguistics commentators, and what results is explication of some of the most pressing Beijing-R questions ever to cross the minds of Olympics spectators. The details are elaborated in the section below.

But I’ve been sitting on that conversation since March. The problem is the trepidation that comes with explaining any phenomenon. The ngr could end up a victim of clinical description via the same process that transformed an epic event in which “the sun…seemed terribly darkened over like sackcloth of hair” into, well, a solar eclipse: neat, but understood. Who doesn’t worry that science is the death of poetry, regardless of how convincing Feynman is?

Even worse, it might turn out the ngr is really nothing after all: a microphenomenon explained by some more important general rule, a motley collection of events not worthy of being grouped together.

But the words of Steven Pinker bring back some hope for the ngr. Paraphrasing Kripke, he offers the tantalizing prospect that an idea, once expressed in a word, actually becomes something and will never be made any less “real” no matter what explanation the people in white coats might give.

Kripke argued for yet another possibility, one that most philosophers had never even considered: knowledge that is a posteriori (discovered after the fact), but necessary. The discovery that the Morning Star and the Evening Star were the same thing (Venus) was a posteriori. But once it was discovered, it was a necessary truth — there is no possible world in which the Morning Star and the Evening Star refer to different things…. Kripke’s argument is an attempt to clarify what we are logically committing ourselves to when we use proper names and names for natural kinds. We are, surprisingly, committing ourselves to a certain class of logically necessary truths.

So maybe it’s saying that the “sublime consonant” can’t be debunked?! I say, therefore it is? Sort of. It seems like the act of naming is a weighty business. Pinker goes on:

Perhaps it is not surprising that people in so many cultures think words have magical powers…, or that one of the gospels should begin, “In the beginning was the Word, and the Word was with God, and the Word was God.”

And that’s exactly what was going on with Osho, although in this case he’s talking about the other side of the coin. Words create meaning, and in some cases that’s undesirable — as in when you can’t express what you’re talking about. It restricts and binds in ways that have profound implications:

Because he has never talked about God, many think Buddha is an atheist — he is not. He has not talked about God because there is no way to talk about God. All talk about God is nonsense. Whatsoever you can say about God is going to be false. It is something that cannot be said.

Other seers also say that nothing can be said about God, but they do say this much, that nothing can be said about God. Buddha is really logical. He will not say even this, because he says, “Even to say that nothing can be said about God, you have said something. If you say ‘God cannot be defined,’ you have defined him in a negative way — that he cannot be defined…’”

Ludwig Wittgenstein, one of the greatest thinkers of this age, one of the greatest of all the ages also, has said, “That which cannot be said must not be said. That which cannot be said, one must be silent thereof.”

Finally, both sides of the issue — that words describe reality while at the same time creating it — are explored by Bohm in his concept of “enfoldment”.

… each perception of a new meaning by human beings actually changes the overall reality in which we live and have our existence — sometimes in a far-reaching way. This implies that this reality is never complete.

Does it converge? You can be the judge of whether I should have torn out a page or two, rolled’em up and smoked’em. But even if it’s all a mirage, we’ve still got those -ngr answers…

Expert Testimony

Since the participants in the Beijing-R discussion were gracious enough to allow reprinting, I’ll try to keep the editorial work to a minimum and simply reorganize the conversation. Comments come from John Wells, Nigel Greenwood, Kwan-Hin Cheung, and Victor Mair and here’s what they had to say about the -ngr and about Mandarin rhoticization in general.

How would you describe -ngr, phonetically?
The starting hypothesis was to characterize it as a “nasalized retroflex approximant.” But after some conversation, the conclusion seems to be that the effect is to produce a nasalized, retroflexed vowel.

Wells: The obvious way to write tangr would be [tʰɑŋɻ], if that is how it is pronounced (i.e. with a sequence of final consonants)… However it may be that the r-effect is merged with the velar nasal to make a single segment. If that is so, perhaps we could write [tʰɑŋ˞], using the IPA rhotacizion diacritic. Or is the result actually a nasalized retroflex approximant? If so, then [tʰɑɚ̃]… A rhotacized ŋ would have contact between the back of the tongue and the velum, with the velum lowered (= the specification for ŋ) plus retroflexion of the tongue tip (which would not have much effect on the resultant sound, since there is no air passing through the mouth cavity). A nasalized retroflex approximant would be a sound something like like AmE ɚ in summer or dinner, i.e. with the soft palate lowered during the vowel because of allophonic assimilation of nasality. I should think this is more likely to be what you’re hearing.

Greenwood: The most concise description I know of this pronunciation comes in YR Chao’s Grammar of Spoken Chinese. In the section “Morphophonemics of the Retroflex Ending” on p46, he has this to say (note that in his GR transcription the retroflex -r is written -l):

With the back nasal ending -ng, a compromise is followed by having the preceding vowel both nazalized and retroflexed, thus: sheng + -l ->shengl = [ʂə̃˞]

Do linguists consider -ngr to be distinct from -nr?
This question came up in the Woods, Pears & Jingle Bells guest post back in February. The short answer is yes. Mair quotes a straightforward er-ization rule from memory out of the same Chao grammar…

Mair: What I remember (albeit vaguely) from those far away days is basically this: (V stands for the YUNMU or vowel of the syllable)

V + nr sounds just like V + r
V + ngr resiults in nasalized V + r

And Cheung sends a page from the Encyclopaedia Sinica that comes to the same conclusion. (Note that those are not tone marks over the vowels, but tildas indicating rhoticization)

Is the -ngr “exotic” — i.e. occurring in few other languages?

Wells: It’s an allophone of AmE syllabic r. I think you would certainly expect to get it in a phrase such as minor name, between the two n’s. Apart from Mandarin and American English, an r-coloured mid central vowel is extremely rare in the languages of the world. It just happens to occur in two of the most widely spoken ones. I’m sure that this is true of the nasalized equivalent, too.

Greenwood: I don’t know of any other language that has precisely this combination of retroflex and nasalization.

Any tips for Zhonglish speakers on learning -ngr?

Greenwood: If you had to get across the pronunciation of “tangr” to a non-native speaker of Chinese, I suppose you could suggest trying to say a rhotic (eg General American) “tar” while simultaneously pinching your nose. The trick is then to let go of your nose without losing the twang!

Mair: I have tried to pronounce -ngr just the way it is spelled, but it seems almost physically impossible to really make that velar come out at the same time as one is nasalizing and rhotacizing. Moreover, when I hear Chinese speakers talking, the nazalization of the vowel that results from V + ngr is really quite conspicuous, and the rhotacization is there, but the velar quality is absent, whereas, when I hear Chinese speakers pronounce V + nr, the vowel is not nasalized, but the rhotacization is obvious.

The Ultimate Beijinger

All this talk starts to get uncomfortable — masochistic, nearly. The Beijing-R fetish isn’t meant to be riled up like this; it’s dangerous territory. The double-/z/ opulence of the fricatives in “nasalized” hum like the Siren’s call. Běijīnghuà is beckoning! You catch yourself unconsciously hanging on the initial rhotic of “retroflex”, hoping against logic it might offer a little more than just description…

Never fear. The nasalized retroflexed vowel is close at hand. No talk. All sound. The real thing. Out of Beijing Sounds cold storage we have rescued a recording from Beijing TV. At the time, no one could see what purpose it might serve, but its timely rediscovery will serve to justify the business expense of such lavish archiving facilities.

The recording suffices to prove that not every TV personality has had their Beijing Dialect harmonized to exacting pǔtōnghuà (普通话, i.e. standard Mandarin dialect) standards. Staff analysts claim to have identified a dozen Beijingisms in just the intro (here’s the whole thing, not Firefox friendly). I’ll leave it to the reader to see if they deserve their jobs and simply point out, in the third line below, the ability to turn what might be three syllables in some world ( dàn yǒu rénr = 但有人儿) into one extended rhotic conglomeration. Beautiful.

[Go to website or bottom of this post to listen to audio]

说的一个字儿， 想必大伙儿都很感兴趣
Shuōde yīgè zìr, xiǎngbì dàhuǒr dōu hěn gǎn xìng qu
This one word, presumably everyone has an interest in

就是钱呢。咱北京和这钱有关系的胡同儿，也有这么几条
Jiùshìr qián ne, zán Běijīng hé zhèi qián yǒu guānxi de hútòngr, yě yǒu zhème jǐ tiáo.
It’s money. And in our Beijing there are a few neighborhoods that have a connection with this money.

但有人儿说了，这胡同儿以前肯定是印钞厂
Dàn yǒu rénr shuō le, zhèi hútòngr, yǐqián kěndìng shì yìnchāochǎng
But there are people who have said that these neighborhoods earlier were definitely money printers.

您说的不全对。所以今儿各我跟你聊了北京和钱有关系的胡同儿。
Nín shuō de, bù quán dùi. Suǒyǐ jīnr ge wǒ gēn nǐ liáo le Běijīng hé qián yǒu guānxi de hútòngr.
You said it wasn’t completely right. So today I’m going to chat with you about Beijing’s neighborhoods with money connections