Problem: Laws and Social Norms Regarding Free Speech

Standards for appropriate speech are regulated both by societal norms and, in many cases, by local law. The bounds of free speech vary considerably from country to country. The Internet magnifies the issue because its inherent nature allows speech to flow freely divorced from political boundaries. Although Chapter 18 will cover the Internet and free speech under U.S. law, here we look at the issue from an international perspective, not with regard to the speaker, but rather to the service provider that facilitates transmission of the contested speech. Because service providers such as Google, Facebook, and Twitter store the speech of others on their servers, one issue is whether such a provider can, should, or must remove content that is offensive to portions of its audience, particularly after the provider has been notified of the specific content found objectionable. For example, in 2011 Google received more than 1,900 requests from governments for the removal of various content, many of which were honored by blocking access from those countries under Google's policy of complying with lawful requests. Which country's requests topped the list? Brazil with 418; the United States was second with 279. This is not as surprising as it might seem when one factors in that countries where less free speech is tolerated block content more directly. China's "Great Firewall" is a case in point. What speech draws a red flag in Brazil?

Racial issues, but also some political speech. Brazil "has strict electoral laws that limit criticism of candidates in the runup to elections."11 In Spring 2013, Facebook angered a considerable number of its U.S. constituents with regard to content that glorified violence against women. Groups concerned with women's rights have repeatedly reported offensive content, with mixed satisfaction. Frustrated, they used one medium to get the attention of another: They "asked Facebook users to use [a particular] Twitter hashtag . . . to call on companies to stop advertising on Facebook if their ads [had] been placed alongside" pages with names such as "Violently Raping Your Friend Just for Laughs." In a very short time the effort had garnered 224,000 supporters on a petition at change.org and Facebook publically stated that its "systems to identify and remove hate speech have failed to work as effectively as we would like, particularly around issues of gender-based hate."12 Perhaps even more challenging are requests from governments for service providers to turn over information revealing the identity of the content poster. France has antihate laws that proscribe anti-Semitic speech. Twitter had already removed at least some of the content at issue when a French court in a civil suit brought by a citizen's group ordered Twitter to identify the individuals who had posted the content. Although its policy is to provide such information only in response to a valid order of a U.S. court, ultimately Twitter agreed to provide the information to French authorities that were seeking to prosecute the posters under their antihate laws. In deciding to cooperate, Twitter may have been concerned that the employees of its Paris offices would be subject to prosecution.13

Question

Acceptable speech varies not just by country but also over time. Today's Internet content issues can be seen as a modern version of TV broadcast censorship. In the early years of TV, bedrooms of spouses could be shown only if they contained two beds. TV censorship varied somewhat by the time of day of the broadcast, but language and dress had some restrictions applicable at all times. As discussed in this section, sensitive topics can also be based on local history and conditions. Additional examples include Thailand's prohibitions against insults to its monarchy and India's fears about content that could spark religious violence. Hate speech is widely restricted. Try developing a substantive rule that an Internet provider could use to determine when a request to remove content should be acted upon, regardless of the country of origin. If your rule references hate speech, it will be important to indicate what speech meets the criteria.