Line censoring messages in China

370 banned words found; firm mum on who's selecting them and why

by Tomoko Otake

Staff Writer

In another example of the dilemmas facing Internet companies operating in China, Japan-based instant messaging app provider Line Corp. has been censoring chats among users there, blocking the transmission of politically sensitive words and phrases.

Line, which announced Monday that the number of people using its app has topped 300 million worldwide, launched its services for Chinese users last December under the brand Lianwo, in partnership with Chinese IT company Qihoo 360 Technology Co.

Recent research by the Citizen Lab, an Internet security and human rights research group at the Munk School of Global Affairs at the University of Toronto, has confirmed that censorship functions are activated on Android smartphone and tablet users in China, and that similar functions probably exist for iPhone users as well.

“When you sign up for Line, you need to verify your phone number,” Seth Hardy, a senior security researcher at the lab, said. “The phone number is used to determine what your home country is. If your home country is China, it will activate the censorship functions no matter where you physically are.”

A detailed analysis by the lab shows that censorship currently covers 370 keywords in the Chinese language, ranging from direct and indirect references to former Premier Wen Jiabao and various phrases linked to Falun Gong, a spiritual movement banned in China, to specific names of anti-government activists.

In the case of Lianwo users, the app automatically downloads a list of censored words from Line’s Website (line.naver.jp) so that their messages are checked within each device, but “it is unclear who decides what keywords are blocked and how this process operates in practice,” said Masashi Crete-Nishihata, research manager at the Citizen Lab.

According to the university group, when the “bad words” are detected, one of two things happens. If the sender’s program is registered to a Chinese phone number, it will show up as an error and not send any message that contains the blocked words, the lab says. If the recipient is being censored but the sender is not, the message will be sent but with blacklisted words replaced with a series of asterisks.

Hazuki Yamada, the global PR manager for Line Corp., acknowledged in an email to The Japan Times that the company “provides services in accordance with the local environment and government regulations of mainland China,” and that therefore “there are occasions when keywords found in violation of Chinese laws are not transmitted.” She declined to offer details on how the banned words are selected.

The company claims, however, that such words are blocked on the devices themselves, and that the blocked exchanges are not sent to Line’s servers. Users’ “communications are not monitored, perused or recorded” by the company, it says.

Yamada added that Lianwo is “different from global versions of Line,” noting that users other than Lianwo are unaffected by the practice.

Reports surfaced recently that in August the Royal Thai Police planned to surveil Line conversations “to ensure the rule of law, order and national security,” following the opening of a police investigation into four people allegedly planning a military coup. Yamada denied that such a “request was made directly by the Thai government” and that the company has not censored messages of users in Thailand.

The Citizen Lab says that while censorship appears to be limited to Chinese users for now, Line certainly has the ability to apply the censorship technology elsewhere.

“Like any messaging service like this, Line has the ability to monitor or censor any messages as they pass through their service’s computer system,” the lab says. “It does not appear as if they are censoring anything else (than in China), but it is difficult to tell for sure.”

Line isn’t the only IT firm grappling with online censorship in China. In 2010, Google withdrew from the mainland and moved its servers to Hong Kong to protest the Communist Party’s censorship policies. But the search giant reportedly made a major concession in January when it dropped a warning message shown to Chinese users when they search for politically sensitive phrases.