Nèige (那个) disfluency and accommodation
In the States I once worked with a group of programmers who were all (mainland) Chinese. During one of their daily arguments over some point of software architecture, a colleague (with zero background in Mandarin) asked me: “So what does ‘nèige’ mean?”
As probably everyone reading this blog knows, it literally means “that” or “those” but it’s far more commonly used as a parallel to “uh” or “well”. In linguistic terms it’s a disfluency, a filler — something you say when you’re thinking about what to say next.
Since we use fillers a lot, it’s funny but maybe not all that surprising that nèige is one of the first individual words that a non-Mandarin speaker would pick out of a conversation that is otherwise, to them, just a stream of sound.
It makes me feel a little better about my own astounding disfluency. On my most recent Beijing Sounds post I noted that what I said during the first few seconds of speech with a cab driver seemed to contain a huge number of neiges. Now that I count them up, it’s even worse than I thought:
Total: 70 words (at least the way I split them up in the Pinyin)
Nèige: 10 instances (14%)
Ouch.
But now Mark Liberman’s Language Log post, “Uh accommodation” gives me hope, that if I can’t learn to speak with fewer nèiges, at least I might (subconsciously) persuade my conversation partners to be disfluent with me.
How does that work? Liberman takes the linguistic idea of “accommodation” — that speakers adapt to the way their conversation partner is speaking — and tests the hypothesis that speakers accommodate “uh”. The findings suggest that accommodation is widespread. Although I’m not a fan of his graphs, the numbers are mind-blowing:
And some ways of looking at the relationship are more striking. Thus of the 4,235 conversations in which the A side never used uh, in 3,507 cases (83%) the B side also never used uh. In contrast, in the 7,464 conversations in which the A side used uh at least once, there were only 752 (10%) in where the B side failed to use uh.
This would be a great thing for some intrepid linguistics student to test in Mandarin. All you’d need would be a good corpus. Speak up if you have one.
Liberman’s own LDC has a fairly large Mandarin corpus, but I think it’s priced somewhere slightly out of our range.
It strikes me that an absolute count of 那个s is probably not that helpful — after all, there are people who will say 那个 three times per disfluency (“我本来想找那个，那个，那个，什么玩意儿来着…”).
”我本来想找那个，那个，那个，什么玩意儿来着…”
Nice. I’m definitely on that 那个 raised to the 那个 plane.
On a slightly serious note, that makes me think that a really good corpus might actually indicate “disfluency” and so would solve all the problems about counting 那个s and not including the ones that are a regular part of speech.
I find after a round of beers, my 那个’s go down very significantly hahaha
Aaron: you might be interested in Sinoglot’s speech-recognition-based 那个 Counter (Model NG — available in our product catalog). Basically it would work something like this:
1. Use the counter during activities at the bar
2. When 那个 counts drop below a specified threshold (individual-specific), refrain from car-driving, date-proposing, etc. etc.
While strictly speaking, this usage would be considered “off label”, you might find it works well as an inexpensive substitute for expensive and cumbersome breathalyzers.
My 那个understanding has always been that it is not so much of a disfluency (which it of course clearly is) but that it also used as a sustituent e.g. when somebody is stalling his speech looking for the right word and hopes the speech partner will fill him in on the word he is looking for, differing from the use of UH in the sense that it is a clear indicator of “hey I can’t find the word, can you help me out here?” Whereas UH in western speech only in few occasions-when the Uh is followed by a longer pause- serves that function.
As to picking it up by non-mandarin speakers, it is just a frequency effect, people tend to pick up the word that is most frequently used, very similary to PERO in spanish, it is used a lot, so even if you don’t speak spanish you will pick that up very quickly.
Well, if you’re affiliated with a research university, you might be able to get the CALLHOME corpus (the LDC one mentioned by Brendan).
I just had a nege experience myself, was staying with some Taiwanese people in NJ, and there was an American visiting who didn’t speak any Chinese at least knew that it’s a filler word…
I always feel that 2 syllable fillers work better than one syllable ones. Saying uhh a hundred times makes you look stupid, whereas shooting a burst of 3 那个s makes you sound like you are unto something… Well, to me at least.
Interestinly, in Spanish we use the same filler 那个, rendered as “este” or “esto”, as well as the English “uuuh”, pronounced more like “eeeh” with the e in “ten”. But of all the fillers I have tried, I think the triple “nege” has to be my favourite.
@Chris: agree, neige does have that use-case that’s “I can’t remember…” but isn’t that what “uh” is used for in English too? e.g. “I’m looking for that, uh…”
@Chrix: gd pt, maybe I should try just contacting the callhome people and see if they’ll let me in…
@Julen: before you decide that neige can sound good, you should go to that linked recording and listen to my painfully dysfluent utterances. You might become a no-neige man.
@syz: uh, yes that is what i meant, but only if UH in english gets followed by a significantly longer space/pause it has that meaning, while in Chinese I have the feeling the spacing is far less or will be followed up by more 那个那个那个。。。making it fairly clear that one is looking for the word. in that sense I agree with julen that it actually seems like you are on to something. In dutch we would say something like: “what is the word, it is on the tip of my tongue” it serves the same function as what you mention: “avoiding to look stupid..”
In japanese they use: Anoo dragging the ooo so that somebody else would fill in their thoughts.