Recently, the New York Times ran an article about the increasingly tight controls over everything from the internet to the media in China. It starts with this anecdote:
If anyone wonders whether the Chinese government has tightened its grip on electronic communications since protests began engulfing the Arab world, Shakespeare may prove instructive.
A Beijing entrepreneur, discussing restaurant choices with his fiancée over their cellphones last week, quoted Queen Gertrude’s response to Hamlet: “The lady doth protest too much, methinks.” The second time he said the word “protest,” her phone cut off.
He spoke English, but another caller, repeating the same phrase on Monday in Chinese over a different phone, was also cut off in midsentence.
Everyone likes a good censorship test, so it’s easy to understand why people started piling on. Shanghaiist and Shanghai Scrap ran tests to see if they could duplicate the effect, and both found they couldn’t. ChinaHush also noted this, and on other blogs and Twitter the response has been kind of harsh, calling the story “false” and attacking the credibility of its authors.
Now, I love a good Western media thrashing as much as anyone. And a Western media blooper that allows us to shout “protest” into our phones? Now that’s good times.
These “experiments” are all predicated, though, on the assumption that the NY Times article is talking about automated censorship. You say “protest”, and presto, the phone magically hangs up on you. And yes, if the New York Times had reported China was doing that, it would be a load of crap. But that’s not what the story says.
Let’s take a look at the sentence that immediately follows that opening anecdote:
A host of evidence over the past several weeks shows that Chinese authorities are more determined than ever to police cellphone calls, electronic messages, e-mail and access to the Internet in order to smother any hint of antigovernment sentiment.
It seems clear to me that the anecdote was meant to be understood in the context of “authorities [being] determined […] to police cellphone calls,” which is probably exactly what was happening. The anecdote isn’t meant to be evidence of voice-recognizing censorship software, it’s evidence of increased police surveillance of the phone calls of anyone they consider suspicious.
Now, there’s no way to know whether the Times’s contacts would fit this description, because both of the sources mentioned are anonymous. Still, other journalists on Twitter confirmed that the authorities are definitely listening in on some phones. So why is everyone assuming the reporters are just making this whole thing up?
The fact is, we’re all testing for an automated system, but that makes no sense. I can’t even imagine the kind of resources bringing such a system to bear on all mobile phone lines would require, and even if they could, how could it possibly work? Given the diversity of accents and dialects throughout China, not to mention the diversity of “sensitive” words, my guess is such a system would be more or less impossible to make effective. Text filtration is one thing — and we already know for a fact that China Mobile filters texts for sensitive keywords from time to time ((I learned that myself the Liu Xiaobo was awarded the Nobel Peace Prize.)) — but voice filtration is another one entirely. Why bother? The PSB already has lists of their “people of interest”, it’s probably a lot cheaper to monitor their phones manually than it would be to monitor the phones of the entire nation via some crazy software.
Now, is this the clearest New York Times story ever? No. The fact that so many people assumed they meant automatic filtration software is testament to that. Moreover, there’s no real way to be sure it’s true because the sources are anonymous. Probably, there was a better way to start that article.
But with that said, is it fair to call the article “false” or accuse them of poor fact checking because yelling “PROTEST!” into your phone didn’t get you disconnected? No. We’ve all had fun playing with our phones. But let’s call off the witch hunt until we have some actual evidence that they’re making things up.
chinageeksarchive 
Just to clarify: nowhere in my post did I suggest that the article was false, or that it was made up. What I did suggest, and stand by, are two things: a) the anecdote that opens the NYT story was not properly verified, and b) that it suggests a particularly “ex-patty” way – ie, not Chinese way – of looking at China’s censorship environment.
Now, in regard to the first point: there are any number of reasons that the phone could have cut out during that conversation, including a bad connection. If the NYT wanted to show that the reason for the cut-out was the use of the English word “protest,” then it very well could’ve tested that word in a phone call. But, rather than do that, they claim to have translated Shakespeare into Chinese and tested that translation. Now, there are plenty of Chinese words that could’ve been used for ‘protest,’ and not all of them turn up as sensitive, either.
In any case, at a minimum, I’d expect the NYT to try more than once. And, in fact, there’s strong evidence to the effect that they DID try to confirm it more than once. Please note that Jonathan Ansfeld is listed as a contributing reporter at the end of the original NYT story. Now, please refer yourself to comment #10 on my blog, left by Jonathan. It reads: “for the record, the contributing reporter’s own tests comport with yours. regrettably his input on the story made little difference.”
LikeLike
protes
LikeLike
@ Adam: I know you didn’t call it false. What I said was “Shanghaiist and Shanghai Scrap ran tests to see if they could duplicate the effect, and both found they couldn’t,” which I think we can agree is accurate, no?
And I don’t disagree with either of your points, although in a newspaper that caters to Americans, it’s understandable that they would approach things in an expat-y way. My point is just that the tests you (and everyone else) did don’t really have anything to do with the point they were trying to make in the article.
I don’t understand how Ansfield’s tests are relevant either, for the same reason. A billion people could test it, but what we’re talking about is an actual human on the other end of the line, not a machine that is guaranteed to respond the same way every time. Now, I wouldn’t be shocked if someone actually WAS monitoring Ansfeld’s phone, but as you know, censorship rules here are super vague and enforcement can vary widely from person to person, place to place, and day to day.
Again, I agree it wasn’t a great anecdote. But I don’t how anyone could possibly “confirm” the story with “tests”; it’s an anecdote from a source, not something that is repeatable in a lab by anyone. I think it’s actually fair to include the story in the piece — it’s interesting, telling, and connects what’s happening in China to something American readers (kinda) know. Or rather, it would have been a good addition to the piece if it had been framed differently. Unfortunately, the way it was written (or, very possibly, edited) makes it sound like they’re suggesting this happens all the time, and the end result is misleading depending on how you read it.
Anyway, my issue was really with the people who are calling for the reporters’ heads on a platter for lying about China, not with anything in your post.
LikeLike
Fair enough. My problem with the anecdote is that, placed at the front of the story, with no note that it might just have been a one-off thing, is highly misleading. And believe me, people are being misled: I’ve got a pile of new hits to my blog from people googling things like “Shakespeare banned phones China.” I’m not calling for anybody’s head – rather, I think a clarification from the NYT would do it. A comment left on my blog by another reporter, I think, offers the best solution:
“I wish the NYT had simply said, or would simply now say, ‘the episode seems not to have been typical and could even have been purely accidental. But it illustrated a larger pattern…’ Because no one doubts the larger pattern.”
Anyway, I appreciate the post, and thoughtful response. I am, if it hasn’t been said before, a fan of the site.
LikeLike
Hate to belabor this, but Ansfield, via his twitter account, is now publicly in favor of the NYT running a correction on this story. See @jansfield.
LikeLike
Charles, really there is no point in defending the indefensible. I appreciate what you do for the sake of debate, but the NYT story doesn’t make sense.
Let’s suppose for a minute it is true that the anonymous guy in the article had his telephone bugged. In that case the listener’so objective would be to know as much as possible of the plan, not to cut the conversation short in the middle, thereby exposing the whole listening scheme!
This is just the kind of idiotic anecdote that some editors like to add in an article, because it works to attract readers. It is the factoid added for color, it is conveniently impossible to prove, and it serves a purely commercial purpose. Let’s not forget that the NYT, like all the other papers, are in the middle of a fight for survival. Like the Chinese say 兵不厌诈 — “all is fair in war”
LikeLike
NYT is a habitual offender when it comes to China FUD:
– The “Lanxiang Vocational School government hacker central” claim has since been discredited by Roland Soong and The Inquirer.
– The “Aurora malware China code” claim has since been discredited by The Register and a 1988 Novell programming guide (the “Nibble CRC” algorithmm has been around for decades.)
LikeLike
“But let’s call off the witch hunt until we have some actual evidence that they’re making things up.” As in half their election reporting since a certain amiable dunce inherited the place?
LikeLike
The sad thing is that this story will probably go to print, and most people who read this will probably believe it. Numbnuts like Glen Beck will probably run with this story.
LikeLike
Did this guy use the iPhone4? You know, the version that have reception issues. Just wondering as I understand Apple products are very popular in Asia.
@pug_ster: I would think Glenn Beck would be more concerned of the Patriot Act.
LikeLike
@ pug_ster : Agreed. In fact, the reason so many people are fired up is that to some extent, that’s already happened. I don’t know about Glen Beck, but Boingboing has a story about how China has automatic voice censorship software that hangs up whenever you say “protest.”
@ Julen: Not necessarily. Depending on when that happened (and I’m assuming it was around the time of certain flower-related activities) the immediate priority would have been to stop any plans to protest from spreading ASAP. That’s the context I understood it in. And given that they were deleting posts about it left and right online to keep word from spreading, it’s not entirely implausible that that’s what happened with this phone call, assuming the guy on the phone was someone police expected to be involved with protests. Now, is that actually what happened? I agree we don’t know, but it seems if we condemn them for making it up without evidence that they did, aren’t we as bad as they (allegedly) are?
LikeLike
The key to bug people’s wireless communication is so that targets don’t discover they are being bugged. If the Chinese government wants to “warn” journalists they can simply send police over and “invite” the journalists for “tea”. For the Chinese government to go through the trouble to bug a journalist’s telephone and to “warn” that journalist by cutting off the conversation after a “key word” being uttered is stupid and illogical, not to mention technically challenging.
If the Chinese government has already developed system to monitor masses of people, and react certain phrases during the conversation in real time, such system would have already been sold to the US government. The US government has been trying for years to develop a system which monitors wireless communications of suspected “terrorists sympathizers”. Imaging a system where anyone uttering a key word over the cellphone and the system would automatically trace the call. That would be the dream of FBI and NSA. I am pretty certain China does not have this capability yet.
Other than the misleading first few paragraphs the rest of the NYT article is okay and informational. Howeve, you expect that a respected and haughty organization like NYT would at least try to verify its claims before printing. Apparently sensationalism is the only way to sell newspapers nowadays.
LikeLike
@ lolz: Nonsense. Why are you assuming the goal is to “warn” them? Like I said, this presumably happened around the same time as that Jasmine nonsense. It’s very obvious that police priority 1 was to STOP INFORMATION ABOUT THE PROTESTS FROM SPREADING. One way to do that is to disconnect the phone line any time one of your sources says something about a protest. It has nothing to do with “warning” anyone, and I don’t think there’d be much point to trying to secretly bug someones phone…anyone who does anything “sensitive” more or less assumes their phone is being tapped anyway, why bother hiding it?
LikeLike
The NYT added an editor’s note to the story a few hours ago. Here it is, in its entirety:
“Editors’ Note: March 26, 2011
An article on Tuesday about Chinese censorship of digital communications began with a description of two interrupted cellphone calls, which were cited as possible examples of “a host of evidence over the past several weeks” that the authorities were increasing their efforts out of concern that antigovernment sentiment might spread from Arab countries. In one call, a Beijing entrepreneur lost his cellphone connection after he used the English word “protest” twice. In the second, a call was lost after the speaker twice used the Chinese term for protest.
The article did not point out that in both cases, the recipients of the calls were in the Beijing bureau of The New York Times. Because scrutiny of press communications could easily be higher than for those of the public at large, the calls could not be assumed to represent a broader trend; therefore, those examples should not have been given such prominence in the article.”
LikeLike
@ Adam: Ah, thanks. That is much clearer! Given that, it certainly does seem the reporters, or perhaps the editors, exercised poor judgment in the wording and placement of that anecdote.
LikeLike
I can’t tell what can be concluded from the NYT article and subsequent editor’s note but the phone cutting off is consistent with how the internet is censored with pages timing out sometimes rather than a straight forward “violating policies” warning page. And you never know if it’s a technical problem or active censorship. If you make the process of sending/getting the message annoying enough, you forget the message and gripe about something else, like an unreliable phone system.
Also, it’s great how Shanghaiist et al tested the phone but a lot of these thing turn out to be pretty close to singular occurrences since the implementation policy is so fluid (makes censorship activities that much more deniable) so I didn’t expect to see positive results.
And if this sounds paranoid, I just question how open the access to foreign news sites are (probably depends on the site, the mood, the place of internet access, the subject, … fluid)
LikeLike
So it’s not China Mobile? Maybe people should try calling NYT’s Beijing office and experiement.
LikeLike
NYT is also guilty of witholding Raymond Davis’ CIA identity from the public, and was forced to admit that when exposed. Also,
“doubletalking dark side of The New York Times, which uses terms such as “enhanced interrogation techniques” when referring to harsh questioning methods by U.S. inquisitors, but uses the word “torture” when the purported perpetrators are, for example, Chinese”
(ref. Alex Gibny, “Freakonomics”)
In essence, would NYT’s China reporting be any different, had it been stat-sponsored?
LikeLike
Uncovering Spoken Phrases in _Encrypted_ Voice over IP Conversations
http://portal.acm.org/citation.cfm?doid=1880022.1880029
Ask yourself: just how easy would that be on unencrypted networks?
LikeLike
“Why are you assuming the goal is to “warn” them? It’s very obvious that police priority 1 was to STOP INFORMATION ABOUT THE PROTESTS FROM SPREADING. One way to do that is to disconnect the phone line any time one of your sources says something about a protest. ”
Because I don’t think it’s currently technologically possible for anyone to listen to millions of phone calls simultaneously, and drop calls in REAL TIME upon hearing certain phrases. I have worked on projects where I had to architect solution for the largest bank in the US to monitor their call centers. As the result I have some idea the amount of hardware, software, and resources which is required to pull this off. This is very difficult to achieve for even private businesses, and private institutions are typically way ahead of the government when it comes to technology adaption. Also, this kind of technology is highly confidential. I suspect NSA has something like this in order to record conversations of suspected terrorists, but there is no way NSA would blow its cover because this technology is a lot better used to spy on people, to figure out the scope of the dissent. Also, China is not afraid of foreign journalists as much as it is afraid of popular Chinese figures, figures who can speak Chinese and appeal to Chinese. GFW doesn’t even censor english sites all that much.
But let’s just say that I am wrong and the technology is there. If the Chinese government wants to stop protests obviously it would try to drop people’s calls not only when the English term “protest” is uttered but also similar terms in Chinese. Since Chinese voice recognition software is not all that great, I would expect tons of false positives which would result in many people’s phone calls getting dropped at a higher rate. If this is so, you would read about it by now not by political magazines but by technology watchers because there would be a drop in service quality for wireless providers.
LikeLike