The Use of Knowledge in Society 哈耶克
政治经济学批判导言读书笔记(优秀版)word资料
政治经济学批判导言读书笔记（优秀版）word资料《政治经济学批判》导言读书笔记《政治经济学批判》导言开篇就指明了物质生产的重要性。
然而，当时的人们还没有很清晰地认识到政治经济学研究的是在一定社会生产关系总体中活动着的物质生产。
没有认识到生产是的、具体的生产。
人们紧紧把生产局限在特定的时期去进行研究和分析。
资产阶级的经济学家之所以要否定上已有的生产方式为人类社会所做的贡献就是旨在说明资产阶级生产方式的唯一伟大之处，就是在极力的维护资产阶级的生产方式。
马克思针对当时存在这样一种不够客观实际的观点就指出了在研究一个时期的生产方式和经济形式时应该客观地去加以分析。
应该把生产的发展看成是一个具体的的过程。
他在这里批判了那种孤立个人观点的肤浅性。
生产与分配、交换、消费的一般关系在这一部分里面，马克思首先是批判了资产阶级经济学认为的“生产是一般，分配和交换是特殊，消费是个别”这种三段论法的肤浅性。
随后就具体的、分别地分析了生产和消费、生产和分配、交换和流通之间的复杂关系。
（一）生产和消费的关系在生产和消费的关系这一节中，马克思主要阐明了生产和消费的同一性的观点。
即生产直接是消费；消费直接是生产。
（1）在阐述生产直接是消费时，他主要从两个方面进行解释。
第一，个人在生产过程中发展自己的能力，也在生产行为中支出、消耗这种能力。
第二，生产资料的消费，生产资料被使用、被消耗、一部分（如在燃烧中）重新分解为一般元素。
因此，生产行为本身就它的一切要素来说也是消费行为。
（2）在阐述消费直接是生产时，他也是从两个方面进行解释。
第一，生产者通过劳动将其劳动行为主要是指其在劳动过程中消耗掉的体力和脑力物化为劳动成果，即产品。
第二，生产者所创造的物人化。
人们在生产过程中创造出产品，然后这些产品又为人们消费掉来维持消费者最基本的生活和再生产的需要。
除此之外，他还进一步深刻的解释了生产和消费之间的关系。
概括起来有下面几种具体的解释：①生产决定消费，生产什么才能消费什么；②消费同时也反作用于生产，如果没有消费，没有消费对象，那么也就不会有生产这种消费对象的生产了。
哈耶克知识问题中的信息与知识论
① ［德 ］格尔 哈德 ·帕普 克：《知识 问题 及其影 响——序 》， ［德］格尔哈 德 ·帕普 克 编：《知识 、自由与秩 序》，黄冰 源 、冯 兴 元 、赵 莹 、梁 晶 晶译 ，中 国社 会 科 学 出 版 社 ，２００１年 。第 ｌ～３２页 。
哈耶克1936年在经济学与知识之中明确地指出知识分工乃是经济学中真正的核心问题他认为中央计划计算不可行的关键在于计划者无法充分利用分散知识dispersedknowledge或者知识分工thedivisionofknowledge而市场价格机制则是利用个人分散知识的最为有效率的信息系统
◎ 谢 志 刚
关 键 词 哈 耶 克 问题 信 息 与 知 识 计 划 经 济 ［中图分类号］Ｆ０１１；Ｆ０９１．３４３ （文献标识码］Ａ ［文章编号３０４４７—６６２Ｘ（２０１８）０６—００４２—０９
“哈耶克 知识 问题 ”在狭 义上 可 以简单 理解 为 “知 识 在 社 会 中的 应 用 ”问 题 ，也 就 是 在 社 会 经 济 之 中如何有效率地利用个人分散 知识 ； 在广 义上 ，则 代 表 了哈耶 克 以认 知 和知识 为 基本 视角 来考察 个 体 、社会和经济 的方法和思想 ，并包含 了从个体到社 会 多个层面 的知识 构建 、演化和 应用 等子 问题。知 识 问题源 自哈耶克在社会主义经济计算大论战 中对 市场价格机制的辩护和对 中央计划经济 的批判 。此 后知识问题不断拓展 ，特别是 与其 自发秩 序理论 联 系起来 ，具有 了一 般 进化 论 、认识 论等 多方 面 的涵 义 。虽然主流经济学界似乎将哈耶克排除在经济学 者之外 ，但随着人们对 经济学 主流 范式 的反思 和批 判 ，新兴 的经济学方法和理论如演化经济学 、行 为经 济学等却越来越多地与哈耶克思想遥相呼应 。
哈耶克社会中知识利用
要成功地做到这一点，他需要多少知识呢？在超出 他的直接知识范围的那些事件中，何者与其决策直接有 关呢？关于这些事件，他又需要知道多少呢？
这个世界上的任何一个地方发生的事件都会影响到 他本应做出的决策。不过，关于这些事件，他不必知道 则多，更不需要知道其所有后果。为什么在某一特殊时 刻对某种尺寸的螺丝钉的需求相对增加了？为什么纸袋 比帆布袋更容易获得？为什么一时较难以获得某类技术 工人或特殊机械工具？对于他来说，这些并不重要。重 要的是这些东西较其他东西的获得变得更为困难或容易 的程度，或对他生产或使用的各种东西的需要在多达程 度上变得更为急切或不则急切了。重要的是他所关心的 那些东西的相对重要性，改变这些东西的相对重要性的 原因并不是他所感兴趣的，也与他本人所处情况中的那 些具体事情无关。
合理经济秩序问题的特殊性恰恰取决于如下事实：我们 必须利用有关我们的处境的知识，但这样的知识从来也 不会以集中或整体的形式存在，而是人们分散地特有的 支离破碎且常常相互矛盾的知识碎片。因此，社会经济 问题并不仅仅是如何配置“给定的”资源—如果这里的 “给定”指将其给予单个头脑来精心解决这些“数据” 所界定的问题的话。社会经济问题更多地是如何确保社 会的每个成员所知道的资源得到最优使用，将其用于相 对重要性仅仅为个人所知的目的。简单说，社会经济问 题是任何单个人都不可能从整体上占有的知识的利用问 题。
经济学家们越来越倾向于忘记组成整幅经济图画的不断 发生的小的变化。原因之一是他们越来越热衷于统计总 量，而统计总量要比细节变量更具稳定性。然而，总量 的相对稳定性并不能像统计学家们偶尔所做的那样，用 “大数定律”或随机变动的相互抵消来解释。要对付的 因素的个数还没有大到偶然因素产生稳定性的程度。商 品和服务流动的背后是蓄意的调整，是为了适应昨天不 曾预计到的新情况而做出的重新部署，是A的失败导致B 的乘机而入。甚至大规模和高度机械化工厂的持续运转 也主要因为存在一种环境，使其能够获得各种意外必需 品，其房顶上的瓦片、办公用的文具、成千上万种它自 己不能生产的设备部件以与该工厂的运营所需要的方案， 这些都要在市场上随手可得。
解读哈耶克《知识在社会中的利用》
解读哈耶克《知识在社会中的利用》哈耶克可能是20世纪最伟大的古典自由主义学者，他倡导市场经济和自由主义，反对过度的政府干预，强调一种“自然秩序”，让市场机制发挥调配资源的作用，当然，哈耶克绝不是一个无政府主义者。
而哈耶克的这些观点并不是空中楼阁，他是通过严格的逻辑论证来阐述的，《知识面社会中的利用》就很好地体现了他的观点和主张。
《知识在社会中的利用》以自由主义和市场价格机制为基础，文章告诉我们社会的经济问题不只是在于如何分配所“资源”，而是如何确保充分利用每个社会成员所知道的资源，这也是文章标题所预示的问题，接一来作者层层论证，突现大家思想的严谨。
文章分为七部分，从信息的掌握和经济秩序的制定谈起，因此，本文也被认为是信息经济学的源头。
哈耶克认为，如果我们能掌握现有方式的全部知识，那么建立一个合理的经济秩序就显得简单了，然而经济运算所依赖的“数据”（即前面所说的信息）从未这样“赋予”一个能由此得出结的单一头脑，于是问题转变为如何利用分散在社会个体中的知识，其实这也是为后面引出价格机制伏笔。
接下来作者就制定经济秩序和利用社会中分散的知识提出了三种制度的制定，即中央计划、自由竞争（即由分散的个人制订的分散计划，这了是基于分散信息的利用）和垄断。
这三种制度哪一种更好呢，“depens mainly on the question under which of them we can expect that fuller use will be made of the existing knowledge”，这里明显可以看出哈耶克的自由主义观点。
第三部分作者主要从分散知识的重要性出发，制度的制定无疑是信息的掌握和利用，若从中央计划出发，那么势必需要一个由挑选的专家组成的权威机构，这个机构或者掌握有最好的知识，但必然无法覆盖社会信息的全部，这在前面已论述过。
这里哈耶克肯定了某一特定时间和地点的知识，它或者称不上科学，但却在经济运行中起到重要的作用。
哈耶克的知识分工
一、引言哈耶克（F﹒A﹒Hayek，1899-1992）是被公认的二十世纪最伟大的自由主义思想家，在其六十余年的学术生涯中，对各种形式的计划经济、集体主义、社会主义、极权主义进行了不遗余力的批判，是坚定的自由主义捍卫者。
由于对自由市场的坚定信奉和市场经济的深刻洞见，哈耶克荣获了1974年诺贝尔经济学奖。
除了是集众多二十世纪伟大的自由主义者的“朝圣山学社”(theMontPelerinSociety)的创立者外，哈耶克还是数个领域的精神领袖，一生涉猎的科学领域包括：经济学、心理学、政治理论，法哲学等。
并且在所涉的每个领域都作出了卓越的贡献。
在对众多领域作了跨学科的研究之后，哈耶克建立起了一个庞杂而坚固的自由主义思想体系，正如J.Gray所说：“哈耶克的论著阐发了一个思想体系，其抱负之宏大完全可与穆勒（John﹒Mill）（由于翻译原因，后文的约翰﹒密尔指同一个人）和马克思(K﹒Marx)的思想体系相媲美，但是却远不如它们易于受到批判，因为哈耶克的体系乃是一种在哲学上站得住脚的有关理性之范围和限度的观点为基础的……。
”凯恩斯（J﹒M﹒Keynes）说“这个世界其实是由思想统治的”这一观点是非常正确的。
在我国建立法治的市场经济的过程中，了解和研究哈耶克的思想有重要的现实意义。
近年来，国内兴起了一股学习和研究哈耶克的热潮，已翻译出版了哈耶克的几乎所有的重要著作。
本文的目的在于理出其思想体系的基石。
在我看来，作为哈耶克思想体系基础的是他于1936年发表的《经济学与知识》（EconomicsandKnowledge）中提出的“知识分工”的思想。
由于“知识分工”的存在，使人们必然对现实世界中的很多事件处于必然的无知状态。
每个行为主体都只能根据自己的信念和所掌握的信息作决策。
作为传递信息的系统的市场协调着不同的行为主体的行动。
市场和其它的制度要有效的运行和发挥应有的作用，个人必须能自主的做各种决策。
因此，自由是必不可少的。
知识在社会中的运用
哈耶克《个人主义与经济秩序》（邓正来译），三联书店，2003年版。
第四章，“知识在社会中的运用”（第三、六节）。
就此而言，不同种类的知识的地位显然是不同的。
因此，对我们这个问题的回答，将在很大程度上取决于不同种类的知识所具有的相对重要性：是那些较可能为特定的个人所掌握的知识更重要，还是那些我们应当较具信心地期望可以为那些经由适当方式挑选出来的专家所组成的某个权力机构所掌握的知识更重要呢？如果说人们在今天普遍认为后一种知识更重要，那只是因为这样一个事实所致：一种知识，亦即科学知识（scientific knowledge），在当下公众的想像中占据了太重要的地位，以致于他们忘记了这样一个道理，即科学知识并不是惟一与此相关的一种知识。
我们可以承认，就科学知识而言，一群经由适当方式挑选出来的专家也许可以最好地掌握可资获得的所有最佳的知识——尽管这种做法只是把这方面的困难转嫁到了如何挑选专家这个问题上面。
在这个方面，我想指出的乃是这样一个要点：即使我们假设这个问题很容易就可以得到解决，它也只是那个所涉范围更广泛的问题中的一个很小的部分而已。
一如我们所知，在今天，谁要是宣称科学知识不是全部知识的话，那他就肯定会被认为是在宣扬类似于异端邪说的东西。
然而，我们只要稍加思索就会发现，现实生活中无疑还存在着一种极其重要但却未经系统组织的知识，亦即有关特定时空之情势的那种知识（the knowledge of the particular circumstances of time and place）——它们不可能被称为科学知识（也就是一般性规则之知识那种意义上的科学知识）。
正是在这个方面，每个人实际上要比所有的其他人都更具有某种优势，因为每个人都掌握着有可能极具助益的独一无二的信息，但是只有当立基于这种信息的决策是由每个个人做出的或者是经由他的积极合作而做出的时候，这种信息才能够得到运用。
就此而言，我们只需要牢记下述几点，我们就能够理解这个道理了：第一，无论我们从事什么职业，我们在接受理论训练以后都必须学习许多其他的东西；第二，学习和掌握各种特定工作的知识，实际上耗用了我们整个工作生涯中的很大一部分时间；第三，在各行各业中，有关人的知识、有关当地环境的知识、有关特定情势的知识，都是一笔极其宝贵的财富。
金融方面的书籍
金融方面的书籍100本推荐一、华尔街人必读的22本金融佳作1、《聪明的投资者》（The Intelligent Investor）Benjamin Graham。
格雷厄姆专门为业余投资者所著、巴菲特称之为“有史以来最伟大的投资著作”。
2、《怎样选择成长股》（Common Stocks and Uncommon Profits）Philip Fisher巴菲特称自己的投资策略是“85%的格雷厄姆和15%的费舍尔”。
3、《投资价值理论》（The Theory of Investment Value）John Burr Williams4、《非理性繁荣》（Irrational Exuberance）Robert Shiller作者罗勃·席勒也是2013年诺贝尔经济学奖得主。
《纽约客》评论道：《非理性繁荣》不仅预言了市场的衰落、更为重要的是、它对于解读投机性泡沫的产生和持续进行了严谨而有益的尝试。
5、《彼得·林奇的成功投资》（One Up on Wall Street）Peter Lynch彼得·林奇是全球首屈一指的投资专家、本书总结了股票投资的诸多技巧、向广大的中小投资者提供了简单易学的投资分析方法、这些方法是作者多年的经验总结、具有很强的实践性、对于业余投资者来说尤为有益。
6、《与天为敌:风险探索传奇》（Against the Gods）Peter L. Bernstein7、《股票作手回忆录》（Reminiscences of a Stock Operator）EdwinLefevre8、《金融炼金术》（The Alchemy of Finance）George Soros本书是索罗斯的投资日记。
索罗斯也许是有史以来知名度最高和最具传奇色彩的金融大师。
这本书讲述了索罗斯如何分析个股、如何把握市场转变的时机、如何面对不利的市场行情并及时调整对策。
9、《证券分析》（Security Analysis）Benjamin Graham and David Dodd。
弗里德里希·奥古斯特·冯·哈耶克有着怎样的思想？又有着怎样的主张
弗里德里希·奥古斯特·冯·哈耶克有着怎样的思想？又有着怎样的主张本文导读：哈耶克是20世纪学术界对于集体主义的主要批评者之一。
哈耶克相信所有形式的集体主义(即使是那些在理论上根基于自愿合作的集体主义形式亦然)最终都只有可能以中央集权的机构加以维持。
在他的著名著作《通向奴役之路》(1944)和其他作品里，他主张社会主义必须要有一个中央的经济计划，而这种计划经济最终将会导致极权主义，因为被赋予了强大经济控制权力的政府也必然会拥有控制个人社会生活的权力。
根基于早期米塞斯和其他人的著作，哈耶克主张道：在中央计划经济里，某个特定的个人或团体必须决定资源的分配，但这些计划者永远都不会获取足够的资讯以正确的分配资源，这种问题又被称为经济计算问题(economic calculation problem)。
哈耶克认为有效的资源交换和使用只有可能经由自由市场上的价格机制加以维持。
而在1945年的《知识在社会中的运用》(The Use of Knowledge in Society)一书中哈耶克主张价格机制可以用以交流和协调个人的知识，使社会的成员能够达成多样化，借由自发性的自我组织原则来解决复杂的难题。
他创造了交易经济学一词来称呼“自我组织的自愿合作制度”。
在哈耶克的观点来看，国家的主要角色应该是维持法治，并且应该尽可能的避免介入其他领域。
在《通往奴役之路》一书中他主张极权主义独裁者的崛起是由于政府对市场进行了太多干预和管制，造成政治和公民自由的丧失而导致的。
哈耶克也对英国和美国因为受到凯恩斯学派影响而建立的经济控制制度提出警告，认为那将会导致相同的极权主义政府产生—而这正是凯恩斯学派所极力避免的。
哈耶克认为极权主义政权如法西斯、纳粹、和共产主义都是同样的极权主义流派;因为这些政权都试着抹灭经济的自由。
对哈耶克而言他认为抹灭经济的自由即代表抹灭政治的自由，因此他相信纳粹和共产主义间的差别仅仅是名称上的不同罢了。
读哈耶克的36句经典名言
读哈耶克的36句经典名言1. "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."This quote highlights Hayek's belief in the limitations of human knowledge and the importance of spontaneous order in economics.2. "The market system is the only system of economic cooperation that is compatible with individual freedom."Hayek believed that the market system was the most efficient and equitable method of organizing economic activity while preserving individual liberty.3. "Socialism is the abolition of private property."Hayek argued that socialist systems that sought to eliminate private property were fundamentally flawed because they ignored the role of knowledge and incentives in economic decision-making.4. "The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess."This quote highlights the importance of decentralized decision-making and the limits of centralized planning.5. "The price system is a mechanism for communicating information and coordinating the actions of dispersed individuals."Hayek believed that market prices reflected the dispersed knowledge and preferences of individuals, and could serve as a guide for economic decision-making.6. "The most important single central fact about a free market is that no exchange takes place unless both parties benefit."Hayek believed that markets were self-organizing and self-regulating, with competition ensuring that exchanges occurred only when both parties saw value in them.7. "The curious task of economics is to teach us how little we know about what we imagine we can design."In this quote, Hayek emphasizes the importance of humility in economic decision-making, and warns against overconfidence in our ability to control economic outcomes.8. "Individualism...is based on the recognition of the value of the individual, and holds that it is desirable that the individual should be free to choose and to act upon his own beliefs and convictions." Hayek believed in the fundamental importance of individual liberty, both for ethical and economic reasons.9. "The success of a market economy, both in its functioning andin its ability to transform an economy, depends on the existence ofa set of institutions which permit mutually beneficial exchange and the recognition of private property rights."Hayek argued that economic institutions were critical for creating the conditions for economic growth and development.10. "There is all the difference in the world between treating people equally and attempting to make them equal."Hayek was a strong critic of policies aimed at equalizing wealth or income, arguing that they were both unsustainable and incompatible with individual liberty.11. "The conflict between the market economy and comprehensive economic planning is not between two different systems of organization, but between two different kinds of systems of organization."In this quote, Hayek highlights the fundamental difference between decentralized, market-based decision-making and centralized planning.12. "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."Hayek believed that economic planning was inherently flawed because it ignored the dispersed knowledge and incentives that underlie economic decision-making.13. "The liberal is not necessarily averse to coercion or compulsion,but to the use of coercion to infringe on the free choice of the individual."Hayek believed in the importance of using limited coercion to protect individual freedom, while also acknowledging the need for some degree of social organization.14. "The preservation of freedom is the protective reason for limiting and decentralizing governmental power."Hayek believed that limiting the power of government was critical for protecting individual liberty and preventing the abuses of centralized authority.15. "The welfare state is designed to achieve a social result that is incompatible with preserving the freedom of the individual."Hayek was critical of the welfare state, arguing that it was fundamentally incompatible with individual freedom and would lead to economic stagnation.16. "The rule of law means that government in all its actions is bound by rules fixed and announced beforehand."Hayek believed in the importance of the rule of law as a check on arbitrary power and a safeguard for individual freedom.17. "The market order is always in the process of becoming, and becoming otherwise than it was before."Hayek believed that markets were self-organizing and dynamic, constantly evolving and adapting to changing conditions.18. "The ultimate result of shielding men from the effects of folly is to fill the world with fools."In this quote, Hayek argues that interventions aimed at protecting individuals from the consequences of their actions can lead to a lack of personal responsibility and poor decision-making.19. "The peculiar character of the problem of a rational economic order is determined precisely by the fact that the knowledge of the circumstances of which we must make use never exists in concentrated or integrated form but solely as the dispersed bits of incomplete and frequently contradictory knowledge which all the separate individuals possess."This quote highlights the importance of decentralized decision-making and the limits of centralized planning.20. "The curious task of economics is to teach us how little we know about what we imagine we can design."Hayek emphasized the inherent uncertainty and complexity of economic decision-making, and the importance of acknowledging our limits in this regard.21. "The fatal flaw of socialism is its ignorance of the fact that knowledge is dispersed, and that no single central authority can possibly know enough to make optimal decisions for everyone."Hayek believed that socialist systems were fundamentally flawed because they ignored the limits of human knowledge and the importance of decentralized decision-making.22. "The spontaneous order of a free society is superior to the deliberate design of a planned society."In this quote, Hayek highlights the importance of spontaneous order and the limitations of centralized planning.23. "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."Hayek believed that economic planning was inherently flawed because it ignored the dispersed knowledge and incentives that underlie economic decision-making.24. "The welfare state is not really about the welfare of the masses, but about the power of the state."Hayek was a critic of the welfare state, arguing that it was primarily focused on expanding the power and control of the state.25. "The idea of social justice is a dangerous fantasy, as it presumes an ability to engineer desirable outcomes that simply does not exist."Hayek argued that the concept of social justice was illusory and dangerous, as it implied a degree of centralized control andplanning that was incompatible with individual freedom.26. "The curious task of economics is to teach us how little we know about what we imagine we can design."Hayek emphasized the importance of humility and acknowledging our limits in economic decision-making.27. "The market is not a place but a process, a process of discovery."Hayek believed that markets were dynamic and constantly evolving processes of discovery, rather than static entities.28. "A government which sets out to abolish inflation also sets out to abolish capitalism."Hayek believed that attempts to eliminate inflation through government intervention could undermine the market system and lead to problems such as inflationary spirals and economic stagnation.29. "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."This quote highlights Hayek's skepticism about the potential for centralized planning to achieve efficient economic outcomes.30. "The essence of capitalist morality is respect for the value of other people's property."Hayek believed in the importance of property rights as a cornerstone of individual liberty and economic growth.31. "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."Hayek argued that economic planning was often motivated by a desire for control and centralization, rather than a genuine concern for economic efficiency and individual freedom.32. "The control of the production of wealth is the control of human life itself."Hayek believed that attempts to centralize economic decision-making in the hands of the state represented a dangerous threat to individual freedom and autonomy.33. "The problem of knowledge is not how to attain it, but how to utilize it effectively."This quote highlights the importance of knowledge and information in economic decision-making, and the need to use it effectively rather than attempting to attain an unattainable level of perfect knowledge.34. "The curious task of economics is to teach us how little we know about what we imagine we can design."Hayek believed that economic knowledge was inherently dispersedand incomplete, and that attempts to centralize decision-making were ultimately doomed to failure.35. "The great aim of the struggle for liberty has been equality before the law."Hayek argued that the rule of law was critical for ensuring individual freedom and limiting the power of government.36. "The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design." Hayek believed that economics should focus on illuminating our limitations and encouraging humility in decision-making, rather than promoting grandiose schemes for centralized control and planning.。
哈耶克｜知识在社会中的运用（完整版）
哈耶克｜知识在社会中的运用（完整版）本文系哈耶克名篇《知识在社会中的运用》，选自《个人主义与经济秩序》（邓正来编译），复旦大学出版社，2012年，第85-100页。
该文系统论述了“知识的分散性”，在哈耶克的知识论中占有重要地位，故推送全文，供朋友们参考。
转自中华好学者ID: zhonghuahaoxuezhe一当我们努力建构一种合理的经济秩序（a rational economic order）的时候，我们想解决什么问题呢？根据人们熟知的某些假设，这个问题的答案是十分简单的。
假设我们拥有所有相关的信息，假设我们能够从一个给定的偏好系统（a given system of preferences）出发，又假设我们掌握了有关可资使用的手段或资源的全部知识，那么剩下的问题也就只是一个纯粹的逻辑问题了。
这就是说，对什么是可资使用的手段或资源的最佳用途这个问题的答案，已经隐含在我们的上述假设之中了。
解决这个最优问题（this optimum problem）所必须满足的那些条件已经完全设定了，因此我们可以经由数学的形式而得到最佳的陈述；用最简洁的话来说，这就是：任何两个商品或任何两个要素间的边际替换率（the marginal rates of substitution）在它们所有不同的用途中都必定是相同的。
然而需要强调指出的是，这根本就不是社会所面对的那种经济问题。
再者，我们为解决这个逻辑问题而发展起来的经济运算方法也没有为我们解决社会经济问题提供某种答案，尽管这种经济运算方法仍不失为我们在解决这个问题的方向上所迈出的重要一步。
这种经济运算方法的发现之所以无法解决社会经济问题，其原因在于：作为这种经济运算方法之出发点的“数据”或“基据”（datum），就整个社会而言，对于一个能够计算其结果的单一心智来说，从来就不是“给定的”，而且也绝不可能是如此给定的。
合理经济秩序的问题所具有的这种独特性质，完全是由这样一个事实决定的，即我们必须运用的有关各种情势的知识（the knowledge of the circumstances），从来就不是以一种集中的且整合的形式存在的，而仅仅是作为所有彼此独立的个人所掌握的不完全的而且还常常是相互矛盾的分散知识而存在的。
The use of external knowledge in factoid QA
The Use of External Knowledge in Factoid QAEduard Hovy, Ulf Hermjakob, Chin-Yew LinInformation Sciences InstituteUniversity of Southern California4676 Admiralty WayMarina del Rey, CA 90292-6695tel: 310-448-8731fax: 310-823-6714email: {hovy,ulf,cyl}@AbstractThis paper describes recent development in the Webclopedia QA system, focusing on the use ofknowledge resources such as WordNet and a QA typology to improve the basic operations ofcandidate answer retrieval, ranking, and answer matching.1.IntroductionThe Webclopedia factoid QA system increasingly makes use of syntactic and semantic (world) knowledge to improve the accuracy of its results. Previous TREC QA evaluations made clear the need for using such external knowledge to improve answers. For example, for definition-type questions such asQ: what is bandwidth?the system uses WordNet to extract words used in the term definitions before searching for definitions in the answer corpus, and boosts candidate answer scores appropriately. Such definitional WordNet glosses have helped definition answers (10% for definition questions, which translates to about 2% overall score in the TREC-10 QA evaluation, given that as many as a little over 100 out of 500 TREC-10 questions were definition questions).This knowledge is of one of two principal types: generic knowledge about language, and knowledge about the world. After outlining the general system architecture, this paper describes the use of knowledge to improve the purity of phase 1 of the process (retrieval, segmenting, and ranking candidate segments), and to improve the results of phase 2 (parsing, matching, and ranking answers).Webclopedia adopts the by now more or less standard QA system architecture, namely question analysis, document / passage retrieval, passage analysis for matching against the question, and ranking of results. Its architecture (Figure 1) contains the following modules, which are described in more detail in (Hovy et al., 2001; Hovy et al., 2000):•Question parsing: Using BBN’s IdentiFinder (Bikel et al., 1999), the CONTEX parser produces a syntactic-semantic analysis of the question and determines the QA type.•Query formation: Single- and multi-word units (content words) are extracted from the analysis, and WordNet synsets are used for query expansion. A series of Boolean queries is formed.•IR: The IR engine MG (Witten et al., 1994) returns the top-ranked N documents.•Selecting and ranking sentences: For each document, the most promising K<<N sentences are located and scored using a formula that rewards word and phrase overlap with the question and its expanded query words. Results are ranked.•Parsing segments: CONTEX parses the top-ranked 300 sentences.•Pinpointing: Each candidate answer sentence parse tree is matched against the parse of the question;sometimes also the preceding sentence. As a fallback the window method is used.•Ranking of answers: The candidate answers’ scores are compared and the winner(s) are output.Webclopedia classifies desired answers by their semantic type, using the approx. 140 classes developed in earlier work on the project (Hovy et al., 2000). These types include common semantic classes such as PROPER-PERSON, EMAIL-ADDRESS, LOCATION, and PROPER-ORGANIZATION, but also classes particular to QA such as WHY-FAMOUS, YES:NO, and ABBREVIATION-EXPANSION. They have been taxonomized as the Webclopedia QA Typology, of which an older version can be found at /natural-language/projects/webclopedia/Taxonomy/taxonomy_toplevel.html.Figure 1. Webclopedia architecture.2.ParsingCONTEX is a deterministic machine-learning based grammar learner/parser that was originally built for MT (Hermjakob, 1997). For English, parses of unseen sentences measured 87.6% labeled precision and 88.4% labeled recall, trained on 2048 sentences from the Penn Treebank. Over the past few years it has been extended to Japanese and Korean (Hermjakob, 2000).For Webclopedia, CONTEX required two extensions. First, its grammar had to be extended to include question forms. The grammar learner portion of CONTEX was trained on approx. 1150 questions and achieved accuracies of approx. 89% labeled precision and labeled recall (Hermjakob, 2001). Second, the grammar had to be augmented to recognize the semantic type of the desired answer (which we call the qtarget ).Its semantic type ontology was extended to include currently abourt 140 qtarget types, plus some combined types (Hermjakob, 2001). Beside the qtargets that refer to semantic concepts, qtargets can also refer to part of speech labels (e.g., S-PROPER-NAME) and to constituent roles or slots of parse trees (e.g., [ROLE REASON]). For questions with the Qtargets Q-WHY-FAMOUS, Q-WHY-FAMOUS-PERSON, Q-SYNONYM, and others, the parser also provides qargs—information helpful for matching: Who was Betsy Ross? QTARGET: Q-WHY-FAMOUS-PERSON QARGS: ("Betsy Ross")How is "Pacific Bell" abbreviated? QTARGET: Q-ABBREVIATION QARGS: ("Pacific Bell")What are geckos? QTARGET: Q-DEFINITION QARGS: (("geckos" "gecko") ("animal"))These qtargets are determined during parsing using approx. 300 hand-written rules.3.Document Retrieval and Sentence RankingAnalyzing the Question to Create a QueryWe parse input questions using CONTEX (Section 2) to obtain a semantic representation of the questions. For example, we determine that the question “How far is it from Denver to Aspen?” is asking for a distance quantity. The question analysis module identifies noun phrases, nouns, verb phrases, verbs, adjective phrases, and adjectives embedded in the question. These phrases/words are assigned significance scores according to the frequency of their type in our question corpus (a collection of 27,000+ questions and answers), secondarily by their length, and finally by their significance scores, derived from word frequencies in the question corpus.We remain indebted to BBN for the use of IdentiFinder (Bikel et al., 1999), which isolates proper names in a text and classifies them as person, organization, or location.Expanding QueriesQuery expansion comes from two sources and used in different stages. In the document retrieval stage, the highly relevant question terms (identified by CONTEX) are expanded in order to boost recall, for example going from “Russian” to “Soviet” or from “capital of the United States” to “Washington”. In the sentence ranking stage, we use WordNet 1.6 (Fellbaum, 1998) to match expanded query terms. Although these expanded terms contribute to the final score, their contribution is discounted. This application of expansion strategy aims to achieve high precision and moderate recall.Retrieving DocumentsWe use MG (Witten et al., 1994) as our search engine. Although MG is capable of performing ranked query, we only use its Boolean query capability. For the entire TREC-10 test corpus, the size of the inverse index file is about 200 MB and the size of the compressed text database is about 884 MB. The stemming option is turned on. Queries are sent to the MG database, and the retrieved documents are ranked according to their ranking from query analysis. We order queries most specific first, then gradually relax them to more general, until we have retrieved a sufficient number of documents. For example, (Denver&Aspen) is sent to the database first. If the number of documents returned is less than a pre-specified threshold, for example, 500, then we retain this set of documents as the basis for further processing, while also submitting the separate queries (Denver) and (Aspen).Ranking SentencesIf the total numbers of sentences contained in the documents returned by MG is N for a given Boolean query, we would like to rank the sentences in the documents to maximize answer recall and precision in the topmost K << N, in order to minimize the parsing and subsequent processing. In this stage we set K=300. We assign goodness score to a sentence according to the following criteria:1.Exact match of proper names such as “Denver” and “Aspen” get 100% bonus score.2.Upper case term match of length greater than 1 get 60% bonus, otherwise get 30%. For example,match of “United States” is better than just of “United”.3.Lower case matches get the original score.4.Lower case term match with WordNet expansion stems get 10% discount. If the original term iscapital case then it gets 50% discount. For example, when Cag(e) matches cag(e), the former may be the last name of some person while the latter is an object; therefore, the case mismatch signals less reliable information.5.Lower case term matches after Porter stemming get 30% discount. If the original term is capital casethen 70% discount. The Porter stemmed match is considered less reliable than a WordNet stem match.6.Porter stemmer matches of both question word and sentence word get 60% discount. If the originalterm is capital case then get 80% discount.7.If CONTEX indicates a term as being QSUBUMED then it gets 90% discount. For example, “Whichcountry manufactures weapons of mass destruction?” where “country” will be marked as qsubsumed. Normally common words are ignored unless they are part of a phrase in question word order. Based on these scores, the total score for a sentence is:Sentence score = sum of word scoresAt the end of the ranking we apply qtarget filtering to promote promising answer sentences. For example, since the question “How far is it from Denver to Aspen?” is asking for a distance quantity, any sentence that contains only “Denver” or “Aspen” but not any distance quantities are thrown out. Only the top 300 remaining sentences are passed to the answer pinpointing module.The bonus and discount rates given here are heuristics. We are in the process of developing mechanisms to learn these parameters automatically.4.Answer Matching using Qtarget-Specific KnowledgeOnce the candidate answer passages have been identified, their sentences are parsed by CONTEX. The Matcher module then compares their parse trees to the parse tree of the original question. The Matcher performs two independent matches (Hovy et al., 2001; Hovy et al., 2000):•match qtargets and qargs/qwords in the parse trees,•match over the answer text using a word window.Obviously, qtargets and their accompanying qargs play an important role; they enable the matcher to pinpoint within the answer passage the exact, syntactically delimited, answer segment. (In contrast, word window matching techniques, that have no recourse to parse structures, have no accurate way to delimit the exact answer boundaries.)Unfortunately, there are many questions, for which the qtarget (which can be as generic as NP), syntactic clues and word overlap are insufficient to select a good answer. Over the past year we therefore focused on strategies for dealing with this, and developed the following.Expected Answer RangeFor quantity-targeting questions, humans often have a good sense of reasonable answer ranges and would find it easy to identify the correct answer in the following scenario:Q: What is the population of New York?S1. The mayor is held in high regards by the 8 million New Yorkers.S2. The mayor is held in high regards by the two New Yorkers.Even without any knowledge about the population of specific cities and countries, a population of 8,000,000 makes more sense than a population of 2. We mirror this ‘common sense’ knowledge by biasing quantity questions like the one above towards normal value ranges.Abbreviation KnowledgeMulti-word expressions are not abbreviated arbitrarily:Q: What does NAFTA stand for?S1. This range of topics also includes the North American Free Trade Agreement, NAFTA, and the world trade agreement GATT.S2. The interview now changed to the subject of trade and pending economic issues, such as the issue of opening the rice market, NAFTA, and the issue of Russia repaying economic cooperation funds.After Webclopedia identifies the qtarget of the question as I-EN-ABBREVIATION-EXPANSION, the system extracts possible answer candidates, including “North American Free Trade Agreement” from S1 and “the rice market” from S2. Based on the perfect match of the initial letters of the first candidate with the acronym NAFTA, an acronym evaluator easily prefers the former over the latter candidate.Semantic Mark-Up in Parse TreesPhone numbers, zip codes, email addresses, URLs, and different types of quantities follow patterns that can be exploited to mark them up, even without any explicit mentioning of key words like “phone number”. For a question/sentence candidate pair likeQ: What is the zip code for Fremont, CA?S1. From Everex Systems Inc., 48431 Milmont Drive, Fremont, CA 94538.Webclopedia identifies the qtarget as C-ZIP-CODE. To match such qtargets, the CONTEX parser marks up (likely) zip codes, based on both structure (e.g., 5 digits) and context (e.g., preceding state code). Two more question/answer pairs that are matched this way:Q: What’s Dianne Feinstein’s email address?Qtarget: C-EMAIL-ADDRESSS1. Comments on this issue should be directed to Dianne Feinstein at senator@Q: How hot is the core of the earth?Qtarget: I-EN-TEMPERATURE-QUANTITYS1. The temperature of Earth’s inner core may be as high as 9,000 degrees Fahrenheit (5,000 degrees Celsius).Using External Glosses for Definition QuestionsWe have found a 10% increase in accuracy in answering definition questions by using external glosses.Q: What is the Milky Way?Candidate 1: outer regionsCandidate 2: the galaxy that contains the EarthFor the above question, Webclopedia identified two leading answer candidates. Comparing these answer candidates with the gloss that the system finds in Wordnet:Wordnet: Milky Way—the galaxy containing the solar systemWebclopedia biases the answer to the candidate with the greater overlap, in this case clearly “the galaxy that contains the Earth”.Finding Support For a Known AnswerIt seems against all intuition that a question likeQ1: What is the capital of the United States?initially poses great difficulties for a question answering system. While a question like Q2: What is the capital of Kosovo?can easily be answered from text such asS2. ... said Mr Panic in Pristina, the capital of Kosovo, after talks with Mr Ibrahim Rugova ...many American readers would find a newspaper sentence such asS1. Later in the day, the president returned to Washington, the capital of the United States.almost insulting. The fact that Washington is the capital of the United States is too basic to be made explicit. In this unexpectedly difficult case, we can fall back on sources like Wordnet: Wordnet: Washington—the capital of the United Stateswhich, as luck would have it, directly answers our question. Based on this knowledge, Webclopedia produces the answer, represented as a lexical target (LEX “Washington”), which the IR module then uses to focus its search on passages containing “Washington”, “capital” and “United States”. The matcher thenlimits the search space to “Washington”. The purpose of this exercise is not as ridiculous as it might first appear: even though the system already knows the answer before consulting the document collection, it makes a contribution by identifying documents that support “Washington” as the correct answer. Semantic Relation Matching in WebclopediaIn question answering, matching words and groups of words is often insufficient to accurately score an answer. As the following examples demonstrate, scoring can benefit from the correct matching of semantic relations in addition:Question 110: Who killed Lee Harvey Oswald?Qtargets: I-EN-PROPER-PERSON&S-PROPER-NAME, I-EN-PROPER-ORGANIZATION (0.5) S1. Belli’s clients have included Jack Ruby, who killed John F. Kennedy assassin Lee Harvey Oswald, and Jim and Tammy Bakker. [Score: 666.72577; 07/25/90; LA072590-0163]S2. On Nov. 22, 1963, the building gained national notoriety when Lee Harvey Oswald allegedly shot and killed President John F. Kennedy from a sixth floor window as the presidential motorcade passed.[Score: 484.50128; 10/31/88; AP881031-0271]Note: Answer candidates are bold (“red”), while constituents with corresponding words in the question are underlined (“blue”) (/natural-language/projects/webclopedia/sem-rel-examples.html). Both answer candidates S1 and S2 receive credit for matching “Lee Harvey Oswald” and “kill”, as well as for finding an answer (underlined) of the proper type (I-EN-PROPER-PERSON), as determined by the qtarget. However, is the answer “Jack Ruby” or “President John F. Kennedy”? The only way to determine this is to consider the semantic relationship between these candidates and the verb “kill”, for which Webclopedia uses the following question and answer parse trees (simplified here):[1] Who killed Lee Harvey Oswald? [S-SNT](PRED) [6] Lee Harvey Oswald [S-PROPER-NAME](MOD) [7] Lee [S-PROPER-NAME](MOD) [8] Harvey [S-PROPER-NAME](PRED) [9] Oswald [S-PROPER-NAME](DUMMY) [10] ? [D-QUESTION-MARK][1] Jack Ruby, who killed John F. Kennedy assassin Lee Harvey Oswald [S-NP]For S1, based on these parse trees, the matcher awards additional credit to node [2] (Jack Ruby) for being the logical subject of the killing (using anaphora resolution) as well as to node [20] (Lee Harvey Oswald) for being the head of the logical object of the killing. Note that—superficially—John F. Kennedy appearsto be closer to “killed”, but the parse tree correctly records that node [13] is actually not the object of the killing. The candidate in S2 receives no extra credit for semantic relation matching.RobustnessIt is important to note that the Webclopedia matcher awards extra credit for each matching semantic relationship between two constituents, not only when everything matches. This results in robustness that comes in handy in cases such as:Question 268: Who killed Caesar?Qtargets: I-EN-PROPER-PERSON&S-PROPER-NAME, I-EN-PROPER-ORGANIZATION (0.5) S1. This version of the plot to kill Julius Caesar is told through the eyes of Decimus Brutus, the protege whom Caesar most trusted and who became one of his assassins.[Score: 284.945; 93/05/15; FT932-8961]S2. Having failed to prevent Cleopatra’s henchwoman Ftatateeta from killing Pothinus, Caesar lets Rufius—the new governor of Egypt—murder her, before turning his back on the lot of them in a devastating display of political indifference. [Score: 264.30093; 92/02/06; FT921-10331]In S1, the matcher gives points to Caesar for being the object of the killing, but (at least as of now) still fails to establish the chain of links that would establish Brutus as his assassin. The predicate-object credit however is enough to make the first answer score higher than in S2, which, while having all agents right next to each other at the surface level, receives no extra credit for semantic relation matching.Good GeneralizationSemantic relation matching applies not only to logical subjects and objects, but also to all other roles such as location, time, reason, etc. It also applies at not only the sentential level, but at all levels: Question 248: What is the largest snake in the world?Qtargets: I-EN-ANIMALS1. Reticulated pythons are the world’s largest snakes, reaching lengths of up to 36 feet.[Score: 384.42365; 12/08/88; AP881208-0148]S2. The amazing Amazon, the widest, wettest and, so National Geographic now affirms, the longest river in the world (4,007 miles, 51 longer than the Nile), boasts the longest snake the most venomous viper, the biggest rat, beetle and ant, along with razor-toothed piranhas that can reduce a Brahman steer to raw bones in minutes and electric eels delivering 640 volts, enough to drive a Metro-North commuter train. [Score: 291.98352; 02/29/88; AP880229-0246]In the S1, [world] receives credit for modifying snake, even though it is the (semantic) head of a post-modifying prepositional phrase in the question and the head of a pre-modifying determiner phrase in the answer sentence. While the system still of course prefers “in the world” over “the world’s” on the constituent matching level, its proper relationship to snake (and the proper relationship between “largest”and “snakes”, as well as “pythons” and “snakes”) by far outweigh the more literal match of “in the world”. Using a Little Additional KnowledgeAdditionally, Webclopedia uses its knowledge of the semantic relationships between concepts like “to invent”, “invention” and “inventor”, so that in example 209, “Johan Vaaler” gets extra credit for being a likely logical subject of “invention”, while “David” actually loses points for being outside of the clausal scope of the inventing process in the second case.Question 209: Who invented the paper clip?Qtargets: I-EN-PROPER-PERSON&S-PROPER-NAME, I-EN-PROPER-ORGANIZATION (0.5) S1. The paper clip, weighing a desk-crushing 1,320 pounds, is a faithful copy of Norwegian Johan Vaaler’s 1899 invention, said Per Langaker of the Norwegian School of Management.[Score: 381.0031; 10/09/89; AP891009-0048]S2. “Like the guy who invented the safety pin, or the guy who invented the paper clip,” David added.[Score: 236.47534; 07/20/89; LA072089-0033]Question 3: What does the Peugeot company manufacture?Qtargets: S-NP, S-NOUN S1. Peugeot intends to manufacture 10,000 cars there each year.[Score: 360.49545; 10/09/89; AP891009-0048]S2. These include Coca Cola and Pepsico, the US soft drinks giants, Peugeot the French car manufacturer, finance companies GE Capital and Morgan Stanley, Nippon Denro, the Japanese steel manufacturer, and the Scotch whisky maker Seagram and United Distillers, the spirits arm of Guinness.[Score: 323.76758; 93/06/25; FT932-902]In S2, “car” gets credit as a likely logical object of the manufacturing process, and “Peugeot”, being recognized as a “manufacturer”, is boosted for playing the proper logical subject role. This example shows that particularly when the qtarget doesn’t help much in narrowing down the answer candidate space,semantic relation matching can often make the crucial difference in finding the right answer.5. Experiments and ResultsWe entered the TREC-10 QA track, and received an overall Mean Reciprocal Rank (MRR) score of 0.435,which puts Webclopedia among the top performers. The average MRR score for the main task is about 0.234. The answer rank distribution is shown in Figure 2. It indicates that we cannot find answers in the top 5 in about 43% of the cases. Once we find answers we usually rank them at the first place.Analysis of the answers returned by the TREC assessors revealed several problems, ranging from outright errors to judgments open to interpretation. One example of an error is a ruling that the answer to “what is cryogenics?” is not “engineering at low temperature” (as defined for example in Webster’s Ninth New Collegiate Dictionary, and as appears in the TREC collection), but rather the more colloquial “freezing human being for later resustication” (which also appears in the collection). Although Webclopedia returned both, the correct answer (which it preferred) was marked wrong. While we recognize that it imposes a great administrative burden on the TREC QA administrators and assessors to re-evaluate such judgments, it is also clearly not good R&D methodology to train systems to produce answers that are incorrect but colloquially accepted. (Checking whether their knowledge is correct is precisely one of the reasons people need QA systems!) We therefore propose an appeals procedure by which the appellant must provide to the administrator the question, the correct answer, and proof, drawn from a standard reference work, of correctness. The administrator can provide a list of acceptable reference works beforehand, which should include dictionaries, lists of common knowledge facts (the seven wonder of the world, historical events, etc.), abbreviation lists, etc., but which would presumably not include local telephone books, etc. (thereby ruling out local restaurants as answer to “what is the Taj Mahal?”).Webclopedia TREC10 Answer Rank Distribution12345XRank Figure 2. Webclopedia answer rank distribution in TREC-10.6.ReferencesBikel, D., R. Schwartz, and R. Weischedel. 1999. An Algorithm that Learns What’s in a Name. Machine Learning—Special Issue on NL Learning, 34, 1–3.Fellbaum, Ch. (ed). 1998. WordNet: An Electronic Lexical Database. Cambridge: MIT Press. Hermjakob, U. 1997. Learning Parse and Translation Decisions from Examples with Rich Context. Ph.D.dissertation, University of Texas at Austin. file:///pub/ mooney/papers/hermjakob-dissertation-97.ps.gz.Hermjakob, U. 2000. Rapid Parser Development: A Machine Learning Approach for Korean. In Proceedings of the North American chapter of the Association for Computational Linguistics (NAACL-2000). /~ulf/papers/kor_naacl00.ps.gz.Hermjakob, U. 2001. Parsing and Question Classification for Question Answering. In Proceedings of the Workshop on Question Answering at the Conference ACL-2001. Toulouse, France.Hovy, E.H., L. Gerber, U. Hermjakob, M. Junk, and C.-Y. Lin. 2000. Question Answering in Webclopedia.Proceedings of the TREC-9 Conference. NIST. Gaithersburg, MD.Hovy, E.H., U. Hermjakob, C.-Y. Lin, and D. Ravichandran. 2001. Toward Semantics-Based Answer Pinpointing. Proceedings of the Human Language Technologies Conference (HLT). San Diego, CA. Witten, I.H., A. Moffat, and T.C. Bell. 1994. Managing Gigabytes: Compressing and indexing documents and images. New York: Van Nostrand Reinhold.。
关于“法治”与“法制”英译名的探讨
关于“法治”与“法制”英译名的探讨作者：王青来源：《中国科技术语》2016年第04期摘要：“法治”与“法制”虽一字之差，但其内涵与外延均不同，英译时需要仔细斟酌。
文章从词典定义入手，比较“法治”与“法制”、rule of law 与 rule by law这两对词的异同。
在此基础上，提出“法治”与“法制”的英译建议。
关键词：法治，法制，rule of law，rule by law中图分类号：C04；H059；D9文献标识码：ADOI：10.3969/j.issn.1673-8578.2016.04.006Abstract： We think both the connotation and extension of fazhi （法治） and fazhi （法制）are different， and their English translations need to be considered carefully. Based on dictionary definition， we compared the two pairs of words， fazhi （法治） and fazhi （法制）， and rule of law and rule by law. We also put forward suggestions for English translation of fazhi（法治） and fazhi（法制）.Keywords： fazhi （法治）， fazhi （法制）， rule of law， rule by law引言从1978年十一届三中全会提出“健全社会主义民主，加强社会主义法制”的目标，到1999年“依法治国”的基本方略被正式写入宪法，再到2012年十八大对全面推进依法治国做出重大部署，强调把法治作为治国理政的基本方式，中国完成了“法制”建设向“法治”建设的转向。
简评哈耶克对科学主义的批判
在科学主义思潮日益得到人们关注的今天，如何看待科学主义已经成为一种关涉历史与现实的问题，在这种情况下，回顾一下发生在二十世纪四十年代的那场关于科学主义的争论，对于我们今天正确理解科学主义是大有裨益的。
正如法国哲学家德勒兹指出的那样：“有两种思考事件的方式，一种方式是沿循着事件，采撷它在历史中的实现，它在历史中的状态，以及它的式微；另一种方式则是追溯事件，像置身于生成中一样置身其中。
”[1]在这里我们更愿意采取第二种方式，因为只有回到历史本身才能做出客观的评价。
作为科学主义思潮发展历程中的著名的争论，哈耶克所提出的具有鲜明个人特点的论证方式，直到今天我们仍能在支持或反对科学主义者的阐释中见到。
因此，在当前重新审查哈耶克对科学主义的批判，对于我们今天正确看待科学主义也会提供一种很好的视角与借鉴。
一哈耶克对科学主义批判的缘起众所周知，F·A·哈耶克（F.A.Hayek,1899-1992）是二十世纪著名的经济学家、政治哲学家，他一生建树颇多，但是在他所有这些成就中，始终存在一条终生未变的主线，那就是对科学主义以及它的各种衍生后果的批判。
哈耶克对科学主义的批判主要体现在1942-1944年，在经济学杂志上发表的一系列论文，后来这些论文汇编成书，就是著名的《科学的反革命：理性滥用之研究》。
在这部书中，哈耶克系统地论证了他对科学主义的批判，在详细分析他的论证之前，有必要对于哈耶克写作这本书的缘由进行一些分析，只有这样才能更好地理解哈耶克的主题，以及对于理解我们今天遭遇到科学主义能有一个清醒的认识。
在我看来哈耶克之所以对科学主义进行了坚决的批判主要有两个原因：其一，哈耶克的学术传承；其二，时代的挤压。
众所周知，一个人的学术传承极大地影响一个人的思维方式与考察问题的视角，哈耶克作为奥地利学派米塞斯（L·Mises）的学生和学术伙伴，自然在思维方式上受到极大的影响。
自此，他接受了贯彻一生的方法论的个人主义，正如他在1944年的著作中写到的那样：“我所努力捍卫的真个人主义在现代的发展，始于约翰·洛克、尤其是伯纳德·孟德维尔、大卫·休谟，后由经由乔赛亚·塔克、亚当·弗格森和亚当·斯密、以及他们伟大的同时代人埃德蒙·伯克的努力而首次达致了其鼎盛时期。
哈耶克思想简介
哈耶克思想简介哈耶克是20世纪最杰出的自由主义者，也是20世纪罕见的社会科学家。
虽然哈耶克1974年获得了诺贝尔经济学奖，但他的学术贡献却远远超出经济学范围。
他毕生发表了130篇文章和25本专著，涵盖的范围从纯粹的经济学到理论心理学，从政治哲学到法律哲学，从科学哲学到思想史。
而且，更重要的是，在这方方面面，哈耶克绝不是玩票，而是见解卓著。
不过，他的贡献主要体现在于，作为一位自由主义者，坚持和重申了古典自由主义的基本理念。
英国学者伯林的划分。
狐狸多知，而刺猬有一大知。
哈耶克自己也曾划分过两种类型：本专业中的万事通，和困惑型人物。
前者对自己的专业了如指掌，熟悉古往今来各种见解，大多数成功的教师、作家和演说家，都属于这类人物。
哈耶克也将这类人称为记忆型的天才。
困惑型人物则凡事都要自己问个为什么，他们获得知识的过程，是自己思考的过程。
他们记不住别人说了什么，他们总是把别人的知识融入到自己的思考中。
他们总是让六经注我，他们固执己见。
他们从各个方面思考，进行探讨，经常是围绕着一个主题。
哈耶克就是这类天才的典型。
第一部分、哈耶克的思想谱系哈耶克1899年5月8日于出生在维也纳一个知识分子家庭，1992年3月23日去世。
他足够长寿，经历了整个20世纪，经历两次世界大战。
亲眼看到了自己所钟爱的观念，自由主义的衰落，及复兴的全过程。
看到极权主义兴起和崩溃的过程。
一、 哈耶克的生平第一阶段：维也纳时代维也纳哈耶克思想成长的时代，是19世纪、20世纪之交的维也纳。
他曾经参加过第一次世界大战。
战争结束之后，在维也纳大学获得的博士学位（1921-1923）。
从思想、知识的角度看，这个时代的维也纳是维也纳历史上最辉煌的时代。
也是奥地利著名作家茨威格在《昨日的世界：一个欧洲人的回忆》中非常生动地描述了19世纪末到1914年第一次世界大战爆发前那段岁月奥地利、主要是维也纳的气氛。
他在回忆录的第一段话就是“倘若要我今天为第一次世界大战千我长大成人的那个时代作一个简明扼要的概括，那么，我希望我这样说：那是一个太平的黄金时代”。
安德鲁·甘布尔：哈耶克视什么为社会科学的核心
安德鲁·甘布尔：哈耶克视什么为社会科学的核心在他的整个职业生涯中，哈耶克的认识论的普遍倾向是朝向波普尔和休谟，而远离米塞斯和康德，但是对他的许多思想，尤其是对他的知识概念而言，仍然是先验的。
哈耶克认为知识是有限的，这不是通过经验探究得出的，而是因为他在人类心智（mind）和人性的本质上的先验立场。
“人不是天生的聪明、理性和善良，而是必须经过教育才能变得如此”（哈耶克，1982）。
人类的知识是有限的，而且必然如此，因此对社会和经济的理解永远是不完整的。
这种认识论上的悲观主义是哈耶克思想的一个基本特征，尽管让人奇怪的是，他从未把这一认识论扩展到自然科学和人类统治自然世界的后果问题，但在哈耶克对传统的日益崇敬和对政府对个人选择的干预的警告方面，这一认识论发挥了作用。
经济学家和其他社会科学家应该从一开始就认识到，他们正在研究他们永远无法完全了解的高度复杂的现象。
对哈耶克来说，知识的有限是从这双重意义上来说的：个体对其情境的了解有限（尽管他对他的情况的了解比其他任何人都多）和控制这些情境的能力有限（哈耶克 1978），以及对其他行为者所拥有的知识的了解有限，而经济学家对单个个体的所知也了解有限。
哈耶克的认识论悲观主义部分植根于康德式（Kantian）的观点，即人类的头脑无法了解世界的本来面目，只能通过头脑（minds）自身所装备的范畴（categories），但主要是休谟式（Humean）的观点，即人类理性是一种脆弱而有限的工具，它产生的知识总是不完美和不完整的。
从康德的角度，并通过奥地利学派的工作，哈耶克认为所有知识都有预设（presuppositions），理论总是先于经验，因此所有知识都因理论而被知会（informed），并以理论为基础。
社会世界表现为一系列复杂的现象，因为这些现象不是自然现象而是社会现象——它们由许多和我们一样的头脑组成，所有这些头脑都是有限的和不完美的，但都有自己的解释、评价和推理能力。
新制度经济学读后感
[标签:标题]篇一：新制度经济学读书笔记浅析科斯《社会成本问题》科斯在1960年发表的一篇名为《社会成本问题》的文章,从此开辟了从社会成本的角度探讨企业的外部性问题。
文章首先批判了庇古在《福利经济学》的某些观点，他认为单纯的惩罚并不能解决好问题，科斯批判的武器就是产权理论。
（一）交易成本的存在科斯认为经济问题都具备交互性，工厂造成了对居民区的污染，但单方面让工厂迁址或者让居民继续忍受都不是最好的选择，必须跳出单方面的局限，从整体来看待究竟是对工厂损失大还是对居民损害多。
科斯又讲了一个案例，假如农夫与养牛者在相邻的土地上经营，牛多了会践踏农夫种的谷物。
在不考虑交易成本时，如果养牛者赔偿农夫损失，他就会减少牛群规模或者把地用栅栏圈起来，再或者雇佣牧人看住牛群，他必须考虑哪种方式最经济。
由于交易成本为零，因此他会不断地进行选择和调整，直到实现他的产值最大化。
但是考虑到交易成本的存在，很多假设必须重新考量，因为此时双方的总成本或者总收益就发生了变化。
在交易成本大于零的情况下免责的一方不会有任何动作，而要负责赔偿的一方只能改变经营方式减小成本，双方的总成本就发生了变化，由此可知交易成本的重要性。
（二）产权与产值前面已经部分涉及到产值的问题。
科斯认为解决社会问题最理想的状态是总产值的最大化和资源的最优配置。
当交易成本为正时情况就不同了，科斯在文章中假设养牛者完全消除损失的成本为80 美元，农夫完全消除损失的成本为50 美元。
当养牛者不负赔偿责任时，农夫则需要设法消除损失，此时成本为50 美元。
而当养牛者负赔偿责任时，其需要花费80 美元成本。
很明显，若养牛者不负责任，将增加30 美元的产值，反之则将减少30 美元的产值。
我们可以看出产权的界定与法律息息相关，产权的界定不同，产值也会相应发生变化。
科斯在后面的内容中通过对法院审理外部侵害案件情况的介绍和分析, 说明在存在交易成本的条件下, 法律体系可以影响经济体系运转的方式, 因而法律在决定资源如何使用方面起着极为重要的作用。
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
“在场的人”的决策如何符合整个经济系统的变化？
不一定要清楚知道这个世界上的任何一个地方发生的事件，重要的是这些东西较其他东西的获得 变得更为困难或容易的程度（相对重要性）——价格机制
价格机制能够协调不同个人的单独行为，使每个都能够依据自己在某一领域所掌握的信息，通过 价格机制的调节而起作用。
eg：世界某地由于某种原因导致某种原料——例如锡，供应紧缺而价格上升，那么这种情况就会 很自然地通过价格机制扩展到世界各个地方。（不必知道具体因为何种原因变得紧缺）
（eg：船主谋生，靠的是充分利用其远洋轮船的运载能力；房地产经纪人的全部知识几乎总 是关于瞬间出现的机会的知识；从事套利活动的人的收益来自商品价格的地 区差异。）
知识
专家组成的权 特殊情况下的
威机构所拥有
个人支配
个体特殊知识的重要性：
贬低特殊情况知识的重要性是因为人们不重视变化，但经济问题总是且唯一地源于变化。
计划制定者的严重分歧：多显著和频繁的变化才需要对生产计划进行重大调整。 但在一个竞争行业中，生产计划是不断进行调整的。Eg：抑制成本上升（如各种原材料价格）是 经理们的一项需要不断努力奋斗的任务。一个无效率的经理很容易失去带来利润的微小机会。 社会主体则可以根据经济中的变化及时改进自己的计划，如A产品的需求突然间上升（当然，可 能原因不知），则企业会根据这一情况调整生产，而中央计划却很难有这样的灵活性。
快捷和经济：这种价格在扩展过程中可以忽略很多不必要的环节，而结果却达到我们的目的，参 与这体系的每个人只需要掌握很少信息便能采取正确的行动。
价格机制——信息传递机制——知识的经济使用
“如果这种机制是人类精心设计的结果，则这种机制早会被誉为人类智慧的一个最伟大的功绩了。 但不幸的是，它既不是人类设计的产物，受其引导的人们通常也不知道自己为何会如此行事。”到 此，哈耶克的目的达到了，如此高效率地利用社会中分散的知识的一种机制不是人类设计的产物， 它是一种人类偶然的发现。这也是为什么哈耶克如此倡导自由主义了。
• 如何制定计划？（如何利用分散在社会个体中的知识？） • 1.中央计划：即由某个权威指挥分散信息制订分散计划 • 3.垄断
如何制定计划？（如何利用分散在社会个体中的知识？） 取决于以下两种做法中何者较为有望取得成功：
• 一是把所有应该利用但分散在很多人中间的知识交给中央权威来处理 • 二是向分散的个人传递一些额外知识，使人们能够做出相互协调的计划
• 哈耶克对社会主义思想的批判与今天大多数中国人对社会主义的批判思路很不一样。今天大多 数对社会主义进行批评的中国人，多是从社会主义实行以后经济效率不高着眼的。而哈耶克对 社会主义批判的着眼点却是制度形成的机制。哈耶克认为自由社会的制度是一种自发的过程， 从没有人有意去设计整个社会制度。因此，这种制度的形成是千万人互相竞争、互相作用的结 果，所以这种自发形成的制度包含了所有人所知道的互不相同的信息，而竞争过程也使人无法 垄断制度设计和形成机制，因而无法利用这种垄断来损人利己，所以信息的合成过程会将信息 传递中的歪曲降到最小。而社会主义制度是世上第一个由某些思想家为社会总体设计的一种制 度，它不是自发地在一个竞争过程中形成的。因此，它包含的信息量受某个思想家所知信息局 限，不可能是亿万人交互作用、将不同的信息综合使用的。
经济秩序如何建立？信息不完备，每个人掌握支离破碎的信息，如何高效利用？ 中央权威 or 特殊情况中的个人？——经济不断变化，中央权威难以快速搜集、 处理信息，做出反应
但个人信息是有限的、支离破碎的，如何使它符合整个经济系统？——价格机制。
1. 市场机制总体上是有效率的。 2. 但是价格反应出的信息也可能是虚假的。 3. 探讨一个问题时，可以从它形成的机制去深入研究，而不只是讨论它的结果。 4. 个人知识是有限的，支离破碎的，需要多探讨、交流、合作。
——杨小凯
• 如何建立一个合理的经济秩序？ • 掌握全部知识（完备市场信息假设）→任何两种物品或要素在所有不同用途之
间的边际替代率相等→难以实现 • 原因：不可能掌握完备信息。
• 人们分散地特有的支离破碎且常常相互矛盾的知识碎片。因此，《知识在社会中的利用》以自 由主义和市场价格机制为基础，告诉我们社会经济问题并不仅是如何配置“给定的”资源，而是 如何确保社会的每个成员所知道的资源得到最优使用。
知识在社会中的利用
The Use of Knowledge in Society
哈耶克 F. A. Hayek The American Economic Review Sep., 1945
哈耶克：奥地利出生的英国知名经济学家和 政治哲学家，代表作有《通往奴役之路》 《个人主义与经济秩序》等。哈耶克可能是 20世纪最伟大的古曲自由主义学者，他倡导 市场经济和自由主义，反对过度的政府干预， 强调一种“自然秩序”，让市场机制发挥调 配资源的作用。哈耶克于1974年获得了诺贝 尔经济学奖，并在1991年获颁美国总统自由 勋章，以表扬他“终身的高瞻远瞩”。
上面提到的知识就其本质来说，无法以统计表的形式传递给中央权威。因为要得到中央权威所需 要的统计数字，就必须略去事物之间的差异。
如果我们承认社会经济问题主要是要在特殊情况下适应变化的迅速调整问题，那么，最终决策就 必须借助熟悉这些情况的人（在场的人）。
但是，"在场的人"不能仅仅依据有限的和隐情般的特殊情况知识来做出决策。还要解决的一个问 题是，如何向他传递另外一些信息，使其决策符合整个经济系统的变化。