Who controls the conversation 誰來控制談話
How to deal with free speech on social media 如何處理社交媒體上的言論自由
It is too important to be determined by a handful of tech executives
這太重要，以至於無法由少數技術主管決定
Oct 22nd 2020
It is the biggest antitrust suit in two decades. On October 20th the Department of Justice (doj) alleged that Google ties up phone-makers, networks and browsers in deals that make it the default search engine. The department says this harms consumers, who are deprived of alternatives. The arrangement is sustained by Google’s dominance of search which, because of a global market share of roughly 90%, generates the advertising profits that pay for the deals. The doj has not yet said what remedy it wants, but it could force Google and its parent, Alphabet, to change how they structure their business. Don’t hold your breath, though: Google dismisses the suit as nonsense, so the case could drag on for years.
這是二十年來最大的反壟斷訴訟。10月20日，美國司法部（doj）指控Google勾結手機製造商、網路商和瀏覽器商，商議用Google為預設搜尋引擎。司法部表示，消費者權益因此受損，別無選擇。這項商定由Google在搜尋界的主導地位所支撐，因為占全球市場約90％，產生廣告利潤支付交易費用。美國司法部尚未透露想要的補救措施，但可能迫使Google及其母公司Alphabet改變業務結構。別屏息以待：Google對這些控告嗤之以鼻，因此案件可能拖至數年。
Action against Google may seem far from the storm gathering against Facebook, Twitter and social media. One is laser-focused on a type of corporate contract, the other a category 5 hurricane of popular outrage buffeting unaccountable tech firms for supposedly destroying society. The left says that, from the conspiracy theories of QAnon to the incitement of white supremacists, social media are drowning users in hatred and falsehood. The right accuses the tech firms of censorship, including last week of a dubious article alleging corruption in the family of Joe Biden, the Democratic presidential nominee. And yet the question of what to do about social media is best seen through the same four stages as the case against Google: harm, dominance, remedies and delay. At stake is who controls the rules of public speech.
反對Google的行動似乎遠異於集聚在反對Facebook、Twitter和社群媒體的風暴。一個是高度聚焦在公司合約的類型上，另一個則是如五級颶風般的公眾憤怒，連續打擊不負責任、可能破壞社會的科技公司。左派說，從「匿名者Q」的陰謀論到白人至上主義者的煽動言論，社群媒用戶被仇恨和不實訊息淹沒。右派指責科技公司的審查制度，包括上周一篇疑點重重的文章，指控民主黨總統候選人拜登家族的腐敗。至於如何處理社群媒體的問題，最好從與起訴Google案相同的四階段來看：傷害、宰制、救濟和延宕。這攸關公共言論規則由誰掌控。
A tenth of Americans think social media are beneficial; almost two-thirds that they cause harm. Since February YouTube has identified over 200,000 “dangerous or misleading” videos on covid-19. Before the vote in 2016, 110m-130m adult Americans saw fake news. In Myanmar Facebook has been used to incite genocidal attacks against the Rohingyas, a Muslim minority. Last week Samuel Paty, a teacher in France who used cartoons of the Prophet Muhammad to talk about free speech, was murdered after a social-media campaign against him. The killer tweeted an image of Mr Paty’s severed head, lying in the street.
美國人有十分之一認為社交媒體有益，約三分之二則認為引起傷害。自2月來，YouTube已辨識出超過20萬個「危險或誤導性」有關新冠病毒的影片。在2016年投票前，1.1億至1.3億美國成年人看過假新聞。在緬甸，Facebook被用來煽動種族滅絕攻擊羅興亞人--穆斯林少數族群。上周一名法國老師帕蒂用先知穆罕默德的漫畫談言論自由，在社群媒體抨擊後被殺害。兇手在推特上發布被割下帕蒂頭顱躺在街上的照片。
The tech firms’ shifting attempts to sterilise this cesspool mean that a handful of unelected executives are setting the boundaries of free speech. True, radio and tv share the responsibility for misinformation and Republican claims of bias are unproven—right-wing sources often top lists of the most popular items on Facebook and Twitter. But pressure is growing on the tech firms to restrict ever more material. In America the right fears that, urged on by a Democratic White House, Congress and their own employees, the firms’ bosses will follow left-leaning definitions of what is acceptable. Contrast that with the First Amendment’s broad licence to cause offence.
科技公司的權宜之計是企圖移除有害內容，意味一小撮非民選主管，正在設定言論自由的界限。的確，廣播電台和電視台需分擔錯誤信息的責任，共和黨人對偏見的說法未經證實-右翼消息來源經常是Facebook和Twitter上最受歡迎的節目的榜首。但是，對科技公司加大施壓，來限制更多的題材。美國右派擔憂，公司老闆在民主黨控制的白宮和國會及自家員工敦促下，對什麼是可接受的內容將依照左傾的定義。對照下，美國憲法《第一修正案》廣泛包容冒犯性言論。
Elsewhere, governments have also used social media companies to go beyond the law, often without public debate. In London the Metropolitan Police requests that they take down legal, but troubling, posts. In June France’s Constitutional Council struck down a deal between the government and the tech companies because it curbed free speech—an initiative that is sure to be revisited after Mr Paty’s murder. Citing Western precedents, more authoritarian governments in countries such as Singapore expect the tech firms to restrict “fake news”—potentially including irksome criticism from opponents.
其他地方，政府也利用社群媒體公司凌駕法律之上，通常未經公開辯論。倫敦大都會警察要求他們撤下合法但令人困擾的貼文。法國憲法委員會在6月裁決政府與科技公司的一項協議，因遏制言論自由而違憲。在帕蒂遭謀殺後，這項創制權肯定會重新審視。許多專制政府像是新加坡引用西方的先例，期望科技公司限制「假新聞」，潛在包括反對者討厭的批評。
This might not matter were the networks less dominant. If people could switch as easily as they change breakfast cereal, they could avoid rules they dislike. But switching is like giving up your mobile-phone number: it cuts you off from your friends. Social networks have also become so central to distributing news and opinion that they are, says Mark Zuckerberg, Facebook’s founder, a “town square”. If you want to be part of the conversation you have no choice but to be there, soapbox in hand.
如果網路的主導性較小，這可能沒這麼重要。人們若能像換早餐麥片一樣簡單轉換社群，就可以避開他們不喜歡的規則。但是轉換社群就像放棄手機號碼：切斷你與朋友的聯繫。 Facebook創始人祖克伯說，社群網路也成了傳播新聞和輿論的中心，它們是「城鎮廣場」。如果您想參與對話，除了登入別無選擇，直播台就在手上。
This hold over users has one further dismal implication for truth and decency. In order to sell more ads, the tech companies’ algorithms send you news and posts that they think will grab your attention. Political cynics, con artists and extremists take advantage of this bias towards virality to spread lies and hatred. Bots and deep fakes, realistic posts of public figures doing or saying things that never happened, make their job cheaper and easier. They are rapidly becoming more sophisticated.
如此套牢用戶，對真實和合宜性會有更可怕的影響。為銷售更多廣告，科技公司的演算法會發送吸引你注意的新聞和貼文。政治憤世嫉俗者、騙子藝術家和極端主義者，就利用這種對網路瘋傳度的偏愛，傳播謊言和仇恨。機器人程式和深度造假（偽造公眾人物言行的逼真貼文）使他們的工作更省錢省事。
The purest remedy for this would be to change the tech firms’ business model and introduce more competition. That is already working well in other areas of tech, like the cloud. One idea is for people to own their data individually or collectively. The social networks would become utilities paid a flat fee, while people or collectives earned the rent from advertisers and set the parameters for what was served up to them. At a stroke that would align the gains from advertising with the burden upon the people being advertised to. If users could port their data to another network, the tech firms would have to compete to provide a good service.
最單純的補救是改變科技公司的商業模式，並引進更多競爭。在其他科技領域（例如雲技術）已發揮很好地作用。一種想法是讓人們個別或集體擁有自己的數據。社群網絡將變成收取固定費用的公共事業，而個人或集體從（租用其資料的）廣告主賺取租金，並為映入眼簾的廣告訂定規範。這樣一來，廣告的收益便會與承受廣告的負擔一致。如果用戶能把自己的資料遷移到另一網路，科技公司將不得不競相提供優質服務。
The obstacles to this are immense. The tech firms’ value would tumble by hundreds of billions of dollars. It is not clear you own the data about your online connections. You could not migrate to a new network without losing the friends who stayed behind unless the platforms were interoperable, as mobile-phone networks are. Perhaps the authorities could impose less sweeping remedies, such as giving users the right to choose feeds set by a neutral rule, not an attention-grabbing algorithm.
這麼做的障礙很大。科技公司的市值將暴跌數千億美元。你的線上連結資料是否歸你所有，並不清楚。你無法遷移到新的網路，而不失去留下來的朋友，除非平台像行動電話網路一樣可互通。或許當局可採取影響較小的救濟措施，像是授權用戶可選擇訊息串流由某種中立規則設定，而不是由攫取注意力的演算法決定。
The keys to the hype house 通往炒作之家的鑰匙
Such ideas cannot be implemented quickly, but societies need solutions today. Inevitably, governments will want to set the basic rules at the national level, just as they do for speech. They should define a framework covering obscenity, incitement and defamation and leave judgments about individual posts to others. International human-rights law is a good starting-point, because it leans towards free speech and requires restrictions to be relevant and proportionate, but allows local carve-outs.
這些想法無法迅速落實，但是當今社會需要解方。不可避免地，政府想制定全國性的基本規則，就像他們為言論所做的那樣。他們應該界定一個框架，涵蓋淫穢、煽動和誹謗，並將有關個別貼文的判斷權留給他人。國際人權法是一個很好的起點，因為它傾向言論自由，要求任何限制必須相關和相稱，但允許地方因地制宜。
Social-media firms should take those standards as their basis. If they want to go further, attaching warnings to or limiting content that is legal, the lodestars should be predictability and transparency. As guardians of the town square, they ought to open their processes to scrutiny and particular decisions to appeal. Ad hoc rule changes by top executives, as with the recent Biden decision, are wrong because they seem arbitrary and political. Hard cases, like kicking opponents of Bashar al-Assad in Syria off a platform for mentioning terrorists, should be open to review by representative non-statutory boards with more power than the one Facebook has created.
社群媒體企業應以這些標準為基礎。如果他們想更進一步，對合法內容加諸警告或予以限制，應當依循可預測和公開透明的指導原則。身為城鎮廣場的守衛，他們的流程必須開放外界檢視，特殊決定應提供申訴管道。高階主管臨時更改規則--像是最近關於拜登的決定--是錯誤的，因為似乎既武斷又政治化。困難的案件像是稱敘利亞阿薩德政府的反對者為恐怖分子，並踢出平台，應開放給具代表性的非法定委員會審查，並給予比Facebook委員會更大的權力。
Independent researchers need much freer access to anonymised data so that they can see how platforms work and recommend reform. Such rule-making should be open to scrutiny. In America politicians can use removing the protection from prosecution granted by Section 230 of the Communications Decency Act as a lever to bring about change.
獨立調查員需更自由地管道取得匿名數據，以便他們能夠了解平台的運作並提出改革建議，這類規則制定應被公開審查。美國政治人物可移除《通信法》第230條所賦予免於起訴的保護，作為改革的槓桿。
This will be messy, especially in politics. When societies are divided and the boundary between private and political speech is blurred, decisions to intervene are certain to cause controversy. The tech firms may want to flag abuses, including in post-election presidential tweets, but they should resist getting dragged into every debate. Short of incitement to violence, they should not block political speech. Politicians’ flaws are better exposed by noisy argument than enforced silence. ■
這將是混亂的，尤其在政治上。當社會分裂，私人和政治言論的界限模糊時，決定去干預肯定會引起爭議。科技公司也許想舉報濫用行為，包括後選舉時期總統的推文。但他們應避免捲入每一場辯論。只要沒有煽動暴力，政治言論都不該封殺。嘈雜的辯論，比強制消音更能揭露政治人物的缺陷。 ■
沒有留言:
張貼留言