Why groupthink is the genius of the internet
Friedrich Hayek, liberal philosopher and economist, was born in the 19th century. Did he accidentally predict the genius of the internet?
Back in 1973, when not even the average nerd knew about the net, Hayek was writing: "Each member of society can have only a small fraction of the knowledge possessed by all and . . . civilisation rests on the fact that we all benefit from knowledge which we do not possess." That certainly sounds like a manifesto for blogs and wikis and all the other smart collaborative tools of the information society. Like democracy, they are based on the wonderfully egalitarian notion that even the lowliest among us has something useful to contribute. But can that possibly be true?
Professor Cass Sunstein, one of the biggest of America's internet big thinkers, has written an intriguing new book, Infotopia: How Many Minds Produce Knowledge*, in which he argues that Hayek's insights about the genius of markets are equally true of the internet. Each individual has incomplete knowledge about things that really matter - but if we get together and share what we know, we can solve our problems, whether that means setting prices or coming up with other solutions based on massive amounts of widely distributed information. All this has implications for how legislatures make laws - the current battle in the US Congress over broadband regulation has been motivated largely by blogs - but also for some of the most basic principles of American law.
Prof Sunstein argues, for example, that sharing scientific information online would cure some of the worst problems of the US patent system and foster innovation much more efficiently than costly patent litigation. Many companies are already using wikis to get employees thinking together online, with surprising improvements in efficiency.
Nothing is perfect, of course: markets can be distorted by falsehood or manipulation, and so can the internet. And because most blog and wiki participants have no financial stake in the accuracy of the information they convey, the economic incentives that drive the markets usually do not apply online. Information sharing on the internet seems to be driven by an unlikely combination of ego and public spirit: people genuinely want to communicate their knowledge.
The average broadband-connected American adult relies on this aggregated information for an astonishing - if not frightening - array of basic information about daily life. Our household recently had an outbreak of headlice, and I gleaned far more knowledge about louse lifestyle and treatment from Google than from the doctor.
In the dark ages before broadband, I might have tried to tap into the collective wisdom on this scourge of modern life by asking family or neighbours. But instead I mined the global wisdom using Google and discovered, from a UK contributor, the best tip so far: wash the kids' hair in Coke (it helps remove the louse eggs). Why did I not think of that myself?
Well, as Hayek and Prof Sunstein would say, that is the human condition: none of us has all the answers. But some of us are not just uninformed, we are actually wrong. Wikipedia, the free online encyclopedia used by millions of people around the globe, insisted in its lice section that I wash my daughters' hair in dog shampoo. The doctor said that was not a good idea. The genius of Wikipedia is that anyone can edit out such errors. It is not just written by the hoi polloi, it is peer-reviewed by them too. So it is usually remarkably accurate - but that kind of thing does make one wonder.
Prof Sunstein recognises all the potential flaws of such collaborative projects. "For aggregating information, the internet offers great risk as well as extraordinary promise," he writes. In the digital world, obtaining the views - right or wrong - of millions of people is virtually effortless. "Every day, like-minded people can and do sort themselves into echo chambers of their own design, leading to wild errors, undue confidence, and unjustified extremism," he says.
Online electronic prediction markets, for example, in which ordinary people bet on the outcome of various events, were more accurate than professional pollsters at predicting the outcome of the last US presidential election. But when it came to forecasting the identity of President George W. Bush's nominee to be chief justice of the US Supreme Court, for example, the masses were woefully misinformed. Everyone was saying that Mr Bush would choose a woman - so everyone else said so too.
Although 57m American adults read blogs, according to a recent survey from the Pew Internet & American Life project, few would pretend that all blogs are accurate. Some are remarkably good; and others are terrible.
Groupthink can be dangerous. But, says Prof Sunstein, the wisdom of the many is a great thing, and sharing knowledge online can lead to remarkable advances for companies, for governments and for the rest of us.
*Infotopia: How Many Minds Produce Knowledge, Cass Sunstein, OUP, September 2006
问不倒的网上智慧
自
由主义哲学家和经济学家弗雷德里希?哈耶克(Friedrich Hayek)出生于19世纪。但他是否碰巧预言了互联网的精髓所在呢？
早在1973年，人们还不知网络为何物时，哈耶克就写道：“每一个社会成员只能掌握人类全部知识中的一小部分……文明须依赖这样一个事实：我们都能获益于自己未掌握的知识。”毫无疑问，这听起来就像博客(blog)和维基(wiki)，以及信息社会所有其它精妙协同工具的宣言。如同民主社会一样，它们基于完美的平等观念，即我们之中最卑微的人也能做出有用的贡献。但真的可能如此吗？
美国最重要的互联网思想家之一卡斯?桑斯坦(Cass Sunstein)教授，写了一本耐人寻味的新书――《信息乌托邦：众人如何产生知识》(Infotopia: How Many Minds Produce Knowledge)*。他在书中写道，哈耶克关于市场精神的精辟见解同样适用于互联网。每个个体对于重要事物都只拥有不完全的知识，但如果我们联合起来，共享知识，就可以解决问题，不论这意味着定价，还是基于数量巨大、传播面广的信息，想出其它解决方案。所有这些不仅对立法机构如何制定法律具有重要意义（目前美国国会关于宽带管制的争论，在很大程度上受到了博客的推动），而且对美国法律中一些最基本的原则也有启示作用。
举例来说，桑斯坦教授认为，在线共享科学信息将解决美国专利体系中一些最棘手的问题，并且与代价高昂的专利诉讼相比，将以更高效率推动创新。许多公司已开始利用维基，让员工在网上集思广益，结果效率获得惊人的提高。
当然，凡事没有十全十美：弄虚作假和人为操纵可以扭曲市场，互联网也难逃此劫。博客和维基的多数参与者并不能从他们传达信息的准确性上获得经济利益，推动市场运转的经济激励机制通常并不适用于互联网。互联网信息共享的推动力，似乎来自自我与公共精神这一看似不可能的组合：人们真心实意地愿意互通知识。
通过宽带上网的普通美国成年人依赖这种聚集的信息，以获取广泛的日常生活基本信息，其程度已经到了惊人的地步。最近我们家染上了头虱病，而我从谷歌(Google)上搜集到的关于头虱生活习性和治疗方法的知识，要比从医生那里获得的多得多。
在宽带出现之前的黑暗时代，我或许已经尝试向家人或邻居求助，利用集体智慧对付这一现代生活的灾难。但现在我利用谷歌挖掘全球智慧，并从一个英国贡献者那里发现了迄今最有用的小窍门：用可乐给孩子洗头发（可乐有助于去除虱卵）。我自己怎么就没想到呢？
换成哈耶克和桑斯坦教授会说，人就是这样：没有人无所不知。但我们中一些人不仅无知，而且实际上拥有错误的知识。全球数百万人都在使用的免费网上百科全书“维基百科”(Wikipedia)在其虱子条目中主张，我应该用犬用香波给我女儿洗头发。医生说，这不是个好主意。维基百科的特点是，任何人都能通过编辑删除此类错误。它不仅由大众编写，也由大众进行同行评审。因此，它通常十分精确――但有关洗头的信息让人生疑。
桑斯坦教授识别出此类协同项目中的全部潜在缺陷。“对于聚合信息而言，互联网带来了极大的希望，同时也带来了巨大的风险，”他写道。在数字世界，要获得数百万人的观点――正确的或错误的――实际上毫不费力。“每天，趣味相投的人能够、也确实把自己归类，进入他们自己设计的回响室，从而导致巨大的谬误、过分的信心和没有根据的极端主义，”他表示。
例如在网上电子预测市场，普通人就各种事件的结果打赌。在预测美国最近一次总统选举的结果时，它比专业的民意测验专家们更准确。但在进行另外一些预测，比如预测布什总统的美国最高法院大法官提名人时，民众就不幸获得了错误信息。大家都说布什将挑选一名女性――所以身边的人也都这么说。
根据皮尤互联网与美国人生活项目(Pew Internet & American Life)最近的一次调查，尽管有5700万美国成年人浏览博客，很少有人会违心说，所有博客都准确无误。一些博客相当不错，而另一些则很糟糕。
集体思维可能很危险。但桑斯坦教授说，许多人集合起来的智慧是个了不起的东西，同时在网络上分享知识能够给公司、政府和我们大家带来显著的效益。
*《信息乌托邦：众人如何产生知识》(Infotopia: How Many Minds Produce Knowledge)，作者凯斯?桑斯坦，牛津大学出版社(OUP)，2006年9月出版。
