This week in not being evil: “loli” spirited away from Google search results

We now interrupt our regularly scheduled absence of programming with a public service annoucement: Super Google has done it again. Not being evil, that is. This has resulted in a suprising lack of results if you happen to ask the wrong question. Like this:

Googling “lolicon”: about 62 results
Googling “lolicon”: about 62 results

About 62 results about “lolicon”, even though the first page advertises 1.5 million hits or so. Login status, SafeSearch settings, private browsing, etc. do not substantially affect that number, and neither does displaying the “omitted results”. See for yourself. For comparison, Yandex claims around 400,000 answers on that search term, and lets you browse 100 pages of results.

More generally, Google has suddenly stopped providing meaningful results to related search queries (anything that includes “loli”, “lolicon”, the Japanese katakana versions ロリ, ロリコン, and possibly more terms), apparently within the last 24 hours. For example, you cannot find the Japanese Wikipedia page about this subject on Google anymore, unless you look up the unabbreviated, much less used phrase ロリータ・コンプレックス (lolita complex) directly. This has caused a bit of a commotion on my corner of the Japanese Internet, including among people known for their level-headedness, like Nakagawa Yuzuru, associate professor of film studies at the Japan Institute of the Moving Image. And I agree with them that the situation is pretty outrageous.

In English, the terms “loli” and “lolicon” are probably unknown outside of the restricted circle of anime fans and Japanese culture specialists, but in Japanese both (especially the latter) are in common use, including on national television, in the news media1, and in scholarly discourse. The English usage may be different, but in Japanese, ロリコン is what people colloquially call an adult man with a not-so-healthy interest in younger girls (be it benign like Lewis Carroll or somewhat more sinister like Humbert Humbert); e.g. the word is thrown around a lot in discussions of the late French painter Balthus and the major exhibition of his works held in Tokyo these days.2 It also has more technical uses that people write entire books about. Certainly, a small fraction of the occurrences of ロリ and ロリコン on the web appear on pornographic sites, but that's mostly because the JAV industry has coopted the terms to refer to what Americans call teen or coed porn: I don't suppose that's a valid reason to ban the word coed from search results altogether.

But that's not all. Words like ロリ and ロリコン are at the centerstage of intense political and legal debate at this very moment, in connection with ongoing Diet discussions over the reform of the law on child pornography, and a current trial on “CG child pornography”. Although they seem to have failed for now, conservative lawmakers (and their lobbyist friends from groups like the 日本キリスト教婦人矯風会 Japanese association of Christian women for moral reform) have, yet again, pushed hard for outlawing fictional depictions of underage characters in sexual situations as tantamount to actual child pornography (see for example the minutes of the debate in the House of Representatives last Thursday). What Google does by supressing such search terms effectively amounts to stiffling discussions of high public interest right as they hit the news headlines.

I know, Google, I know. It's not censorship, you're not a government body, you're just exercising your free speech by publishing carefully crafted search results. But by the same measure, the similarly high-minded people at Baidu are also exercising their free speech by not showing us results about Tian-an-men, aren't they? And they're not claiming not to be evil either.

Why are you doing it, though? You haven't put out any statement yet. If we google for things like “Google ロリ 規制” (Google loli ban), nothing comes up, not even the many webpages that talk about this very problem. Some people have asked you directly on your forums, to no avail so far. So we're speculating. Did the same people that lobby the government pressure you into taking unilateral action? Did some “concerned citizen” convince you that nobody should get to see stuff they don't like, as they did Amazon? Was there some arm-twisting from those Japanese bureaucrats who have pledged to cleanse the country of unsavory material ahead of the 2020 Olympics?

It seems that Bing is doing it too, so it can't be a coincindence. So you owe your users an explanation. And you owe it to yourself to back off, if not being evil means anything.


  1. Just this week, outlets like NTV have talked of “lolicon material” seized by the Tochigi Prefecture police department in a rather sordid murder case, and this has sparked controversy, as the same words had been used by Tochigi authorities to push the infamous wrongful conviction in the Ashikaga case

  2. Dear readers who happen to be in Tokyo: do yourself a favor and go see it if you haven't already. Only two weeks left! 

2 comments for ‘This week in not being evil: “loli” spirited away from Google search results’.

I think your government is doing this and not Google.

http://i.imgur.com/RVnPAWN.png

Just saying...

Try going to the next page of Google results (or click the "Try for yourself" link I conviently put up there), and see how far you can go.

As I said, many hits are announced on the first page (if you just query the word itself; more precise queries tend to turn up nothing at all), but as soon as you move forward, poof.

Post new comment

The content of this field is kept private and will not be shown publicly. If you have a Gravatar account associated with the e-mail address you provide, it will be used to display your avatar.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.
Syndicate content