論公民與政治權利國際公約下仇恨性言論之管制 －以跨國網路為核心

Abstract

2013年03月01日聯合國人權專家於首次審查我國國家人權報告後，公佈結論性意見，提供許多寶貴建議供我國檢討改進。審查結論第74段特別指出：「《公民與政治權利國際公約》第20條規定，任何鼓吹民族、種族或宗教仇恨之主張，構成煽動歧視，敵視或強暴者應以法律禁止之。因此，專家建議應制定法律使鼓吹民族、種族或宗教仇恨之罪行被納入刑法規範。」國家應就仇恨性言論進行管制已為國際法上之趨勢，然而公民與政治權利國際公約（International Covenant on Civil and Political Rights, 以下簡稱ICCPR）於各締約國脈絡化實踐的過程中，受到資訊時代網際網路之去脈絡化特色影響，產生了許多管制爭議。ICCPR如何看待網路時代下仇恨性言論管制之議題，即值得進一步的研究。 因而，本文目的即是欲討論於ICCPR下如何面對跨國網路仇恨性言論之管制爭議？欲回答此問題必須先探求ICCPR判斷仇恨性言論之管制標準為何（單一標準或多元標準）？並依循此問題所得之結論進一步判斷ICCPR面對網際網路上的仇恨性言論是否採取相同之管制標準？當釐清了ICCPR面對真實世界與網路世界之管制差異，方能進一步分析各締約國要如何管制網路仇恨性言論方能夠符合ICCPR之要求？ 欲達上述研究目的，本文首先從ICCPR之性質出發，確立其在國際法上之地位，並發現仇恨性言論管制於國際法上之管制強度，甚至能達到經由國際刑事法庭審判之程度。因而，世界各國皆不能忽略仇恨性言論管制之議題。 其次，本文從歐美就仇恨性言論管制論辯之相關文獻以及聯合國人權事務委員會對ICCPR第20條第2項之闡釋進行分析，發現憲政價值觀的差異導致學界對於仇恨性言論與否之論爭有相當大的歧異。此差異亦反映在各國於管制仇恨性言論時之制度選擇。不過，聯合國人權事務委員會（United Nations Human Rights Committee，下簡稱委員會）相當尊重各國管制仇恨性言論之歧異，容認其在ICCPR第20條第2項下各國可以有不同的管制模式與密度。 在本文第三部分，處理ICCPR如何處理網路仇恨性言論管制之問題。借用傳播學之理論，本文分析網際網路與仇恨性言論相遇後產生之種種爭議，研究發現網路仇恨性言論管制與真實仇恨性言論管制最大的差別在於對於訊息之理解以及追究發言者之困難。不過，縱然面對此困難，透過聯合國對於網際網路管制之文獻分析，亦可發現ICCPR針對網路仇恨性言論之管制，其實仍是維持多元的管制標準並沒有差別對待。 本文在第四部份處理ICCPR之管制標準於網路世界應如何落實之問題。根據Lawrance Lessig教授之網路管制理論，科技實為維持網路仇恨性言論管制多元性之必要手段。據此，本文以應交由「誰」控制網際網路之運作為標準，提出四種跨國網路仇恨性言論管制模型，分別為：政府管制型、代理型、業者管制型、使用者管制型，四種跨國網路仇恨性言論之管制模式，並得視具體之情形交錯適用。 透過上述步驟分析，本文主張跨國網路仇恨性言論管制之核心在於訊息對於各該國家之具體影響，而非處罰位居國外的發言者。由於ICCPR面對網路仇恨性言論管制之爭議是採取多元標準，故各國可依其需求透過科技，針對不同類型的仇恨性言論，以不同管制密度阻絕跨國網路仇恨性入侵。然而，針對網際網路進行結構控制，具有相當強大的管制效果，因此在管制時，毋寧是注重網際網路之特性，散佈更多的言論矯正仇恨性之不良影響，而非是一網打盡之一概禁絕。

Regulating hate speech has become an international consensus, but it is facing with some problems in this information era. The International Covenant on Civil and Political Rights (hereinafter “the ICCPR”) article 20 (2) asks countries to enact laws to prohibit hate speech. While countries are trying to set up the regulations to comply with the ICCPR, there are some difficulties regarding hate speech on the Internet. The purpose of the thesis is to analyze how to regulate Internet hate speech under the ICCPR. For one thing, the standard of hate speech may vary from country to country. What standard should the countries use to define hate speech, when such speech is spread via the Internet? Should ICCPR adopt a universal standard or a differentiated standard? Moreover, even regarding it as hate speech, how can countries regulate or punish the speakers abroad? All these questions are unavoidable, when countries want to enact such laws. To achieve the above, this thesis is divided into four parts. First, the thesis analyzes the trend in international law of regulating hate speech. This thesis finds that a lot of international covenants regard spreading hate speech as a crime, and it is punishable. However, the ICCPR provides countries with multiple measures to regulate hate speech. Second, the thesis tries to find the ICCPR’s interpretation in applying article 20(2). By reviewing scholar’s discussions, analyzing the Human Rights Committee of ICCPR’s (hereinafter “the Committee”) explanation and observing the hate speech law of countries, the thesis finds the following tendency. The ICCPR respects nationalized cultures and decisions to determine what kind of speech is hate speech and whether it shall be punished or not. Therefore, it is possible that the same speech may be considered as hate speech, while in another country does not. Thirdly, the thesis analyzes whether the Committee should apply the same standard when regulating internet hate speech. The thesis analyzes the feature of Internet hate speech, and finds that there are two challenges to regulate it. The first thing is that since the Internet has the characteristic of “de-contexualization”, it is hard to judge whether an internet speech is a hate speech. Secondly, it is also hard to punish speakers who spread hate speech abroad. This thesis analyzes the ICCPR’s interpretation about freedom of the Internet, and finds that the Committee still adopts the same standard to deal with internet hate speech as with real world hate speech. Finally, the thesis analyzes how the ICCPR’s standard is practiced on the Internet. The thesis refers to Lawrence Lessig’s method to analyze the issues, and holds that technology can help countries to efficiently regulate Internet hate speech. Using this method, the thesis concludes that the key to fulfilling the ICCPR’s standard is to block hate speech from other countries, not to punish those who spread it. Moreover, by deciding who has the authority to control the Internet code to block information, the thesis establishes four modes to regulate Internet hate speech, which are the governmental mode, the semi-governmental mode, the ISPs-mode, and the user-mode. The thesis holds that the government-mode is the most powerful method to control the Internet. Nevertheless, by using this method free speech will likely be infringed, hence it must not to be adopted to regulate Internet hate speech. Apart from this, countries may adopt the other three modes to regulate Internet hate speech. In conclusion, this thesis holds that the main point of regulating Internet transnational hate speech is the relationship between technology and messages. Via technology, different countries can have different standards toward regulation of internet hate speech which can be tailored to their own cultures. Hence, the regulation object will shift from the punishment of the speakers to the message itself. Eventually, it will conform to the ICCPR’s “differentiated standard”. However, since technology is a powerful tool, countries should show self-restraint in order not to infringe free speech.