yet more on “chink”

In a previous post, I wrote about my searches on Google’s ngram viewer on the relative historical use of the terms “chink” and “Chink.” Ngram viewer’s dataset only goes to 2008/9, but another Google web tool, trends, tracks the text of web searches (at Google, obviously) since 2004 – yes, trends is a text search of text searches, so for those folks who love self-referentiality, it is in effect a meta-search engine. Since they draw on different data, actual print usage versus web search text, the two tools aren’t directly comparable. Nevertheless, a quick check on Google trends for the term “chink” was interesting.

Google trends normalizes the graph of its search results to the highest peak reached in the time frame selected. My search on “chink” and “Chink” returned identical results – best seen by this search – leading me to think that Google trends, unlike Google ngram viewer, is not case-sensitive. The most apparent result was the trend on the term was mostly steady and low-level with the exception of, and relative to, one significant peak in February 2012. That one peak generated more than ten times the usual volume of searches; in other words and ironically, the controversy over ESPN’s headline during the Linsanity pushed interest in, searches for, and use of the term “chink” on the web to its all-time high on Google trends. It isn’t yet possible to compare Google trends to ngrams directly, but ngrams allows downloads of its data, so it may be possible at some later point. But it could well be that the peak in Google trends may also eventually be significant within the grams viewer. How ironic would it be that the discussion about “chink” in the ESPN headline turned out also to be the most significant instance of the term’s use historically?

Google trends includes an option to link news headlines to peaks in its searches. Those linked to the other peaks from my search were interesting in several ways. First, the majority used “chink” in its older, non-racial sense, usually within a reference to the expression “chink in the armor.”

One headline and search peak, however, from October 2009, involved an instance very similar to the ESPN incident. A player turned commentator on Britain’s BBC Two television channel, John Virgo, also use the expression “chink in the armor” while referring to a Chinese competitor in a broadcast of the Snooker Grand Prix final. As reported in the Telegraph, in lamenting the shortcomings of Ding Jinhui, Virgo said “I think there’s a chink in his armour.” While snooker, a form of billiards, isn’t basketball and Ding not as well-known as Lin, both instances involved sports media coverage — some might argue snooker isn’t a “sport” per se – where the phrase was used in reference to a competitor/player of Chinese descent. In both instances, subsequent reaction, discussion, and debate centered on whether the use was a) racial, b) intentional, c) derogatory, and d) appropriate. In Virgo’s case, the BBC defended him issuing a statement that “This phrase is commonly used and was appropriate in the context of Ding’s play. There was absolutely no intention whatsoever to cause any offence.” In ESPN’s case, Anthony Frederico, the man who wrote the web site headline, was fired and anchor Max Bretos, who had also use the phrase in an on-air interview, was suspended for thirty days. Both apologized to Lin and stated that their use of the phrase was not intended to be racial or derogatory.

Another interesting result from Google trends was the geographic/regional distribution of the web searches it tracked. The vast majority of searches on “chink” occurred in the United States and Canada with the U.K. and Australia third and fourth. From January 2004 to January 2012, searches in Canada actually outnumbered those in the United States slightly (by 3%).From March to December 2012, the U.S. and Canada reversed positions, but again, differed by less than 10%. For February 2012, however, Taiwan ranked as country with the most searches on “chink” followed by the U.S. (at 87% on a relative scale) and Canada (47%). The numbers from February were significant enough to skew the overall regional distribution results, not only for the year 2012 but the entire period from 2004 to 2012. For 2012, with U.S. searches set to 100 on a relative scale, Taiwan followed closely with 96% of the U.S. total followed by Canada (68%), the Philippines (55%), and Australia (27%). For the entire history of Google trends, the figures were U.S. 100; Canada 84; Hong Kong 76; Taiwan 73; Singapore 71; Philippines 52; Ireland 47; U.K. 44; Australia 37; Malaysia 33. The sudden appearance of Hong Kong and Singapore suggests that Google trends’ algorithm to include city-states in its regional distribution needs a bit of tinkering, but the overall result is clear. With the exception of the Linsanity, web searches on “chink” were largely confined to North America. On the other hand, the Linsanity’s size and extent in southeast Asia are also evident. That Lin is enormously popular in Taiwan – whose political status was mostly step sided in discussions about Lin’s ethnic/national heritage – is also made clear while raising questions why Google didn’t track a similarly large surge in trending in China.