长文慎入。

charpter 8 The basic problem: “Mine is better”

提到了一个自相矛盾的点：

People profess enthusiasm for personal growth and development and spend billions of dollars on self-help books, taps and seminars, yet they act as if their minds have no need of improvement.

为什么会这样？

每个人或多或少都会有“mine is better”的想法。

It is natural enough to like our own possessions better than other people’s.

many people realize that we all see ourselves in a special way, different from everything that is not ourselves, and that whatever we associate with ourselves becomes part of us in our minds.

some people think that “mine is better” is not an attitude that everyone has about his or her things, but a special higher truth about their particular situation. 心理学家将这一类人分为两种：个人中心主义者egocentric，种族中心主义者ethnocentric people。

Egocentric People【以自我为中心，个人主义者】

Egocentric means：centered or focuses on oneself and interested only in one’s own interests, needs and views.

太专注于自我，以致于对于这类人来说，成为一个合格的critical thinker是一件很困难的事情。

The perspective of egothink is very limited, egocentric people have difficulty seeing issues from a variety of viewpoints.

Ethnocentric People【种族中心主义者】

Ethnocentric means excessively centered or focused on one’s group.

What distinguishes ethnocentric individuals from those who feel a normal sense of identification with their group is that ethnocentric people believe that their group is not merely different from other groups but foundamentally and completely superior to them and that the motivations and intentions of other groups are suspect.

种族主义者大抵都带有种族歧视的毛病。

[种族中心主义者的这种偏见，会有一个额外的作用。对于社会中存在的各种真实或者想象的问题，他们找到了一个背锅的对象。]

It fills their need for an out-group to blame for real and imagined problems in society.

And they get a large target at which they can point their anger and fear and inadequacy and frustration.

Controlling “Mine is better” Thinking

上面提到的两种类型是“mine is better”的极端形式，但是每个人都会或多或少带着两者的影子，特别是前者egocentric。

“Mine is better” thinking is the most basic problem for critical thinkers because, left unchecked, it can distort perception and corrupt judgement. The more mired we are in subjectivity, the less effective will be our critical thinking.

这里有一点需要弄清楚，并不是一味的否认“mine is better”，而是说“mine is better”的这种思维模式会让人对所有其他的想法弃之不理，想当然的以为自己的就是最好的。但也有可能，TA的这个想法确实是最好的。

回到主题中，作者给到了两种controller的方法：

• keep in mind that, like other people, we too are prone to it and that its influence will be strongest when the subject is one we really care about.
• be alert for signals of its presence, those signals can be found both in feelings and thoughts:
  • in feelings: very pleasant， favorable sensations; the desire to embrace a statement or argument immediately, without appraising it further. or very unpleasant, negative sensations, the desire to attack and denounce a statement or argument without delay.
  • in thoughts: ideas such as “I’m glad that experts are taking such a position - I’v thought it all along and no use of wasting time analyzing this evidence – it must be conclusive”, or ideas such as “this view is outrageous because it challenges what i have always thought – I refuse to consider it”.

charpter 9 errors of perspective

开篇给了一个很好的类比：

当你戴着一副严重变形的眼镜去观察这个世界时，你将自己的发现与身边的人分享，你会惊讶你们看到的世界竟如此不同，困惑于他们没有能力像你那样可以看到如此清晰的世界，这时，你可能会采取如下的行为：停止与周围人的交流，坚信自己的所见，相信你们之间的分歧是他们的问题。然后，在某些偶然的机缘下，你突然发现问题不在他们，而是你佩戴的眼镜，你去找最近的配镜师，换上新的眼镜，发现此时你看到的世界才更清晰。

错误的观点就好像严重扭曲的镜头，只是他们不是架在我们的鼻梁上，而是居住在我们的思想中。

Errors of perspective are like seriously distorted lenses, except instead of being perched on our noses, they inhabit our minds.

They shape the attitudes and habits you bring to the evaluation of issues and create expectations that bias your thinking. Moreover, you may not even be aware of their existence unless you evaluate your patterns of thought.

those errors that create obstacles to critical thinking before we address any issue，具体都有哪些呢？这一章主要提到了7个：

• poverty of aspect
• unwarranted assumptions
• the either/or outlook
• mindless conformity
• absolutisim
• relativism
• bias for or against change

下面一一详细说明了每一项错误的定义，弊端及改进方法。

Poverty of aspect

Karl Duncker, coined the term poverty of aspect to refer to the limitation that comes from taking a narrow rather than a broad view on problems and issues.

这里作者提到上个世纪中，学术在历史进程中的倍增和每个学术领域的知识爆炸，是引起poverty of aspect的不可忽视的原因。

the multiplication of the academic disciplines over the course of history and the explosion of knowledge that has taken place in every discipline, especially during the previous century.

因为领域划分越来越细，专才越来越多，全才越来越少。

这种专注有利有弊：

利：This specialization deepened understanding and multiplied scholarly insights.

弊：It cut off many scholars from the insights of disciplines other than their own and aggravated the condition called poverty of aspect . This poverty creates significant problems in the analysis of complex issues.

策略：

recoginze the limitations of your experience and discipline your mind to broaden your outlook beyond the familiar, to examine all relevant points of view, and to understand before judging.

「可以这样理解，学术或者某个领域的广度和深度，深度视角虽然让人对某个专属领域理解更深，但是会带来poverty of aspect, 或者说tunnel vision，看不到问题的全貌。」

Unwarranted assumptions

这部分篇幅很长，但是内容很赞，提到的很多例子也很有实用。「感觉自己可以对号入座了」

assumption是什么？

Assumptions are ideas that are merely taken for granted rather than produced by conscious thought.

making assumptions是很自然的事情，很多假定不仅无害还有益，比如你早上起来去学校，你假定手表是正常的，校车是正常发出的，教授会按时出现在教室，过马路的时候，你假定汽车是遵守交通规则的，绿灯的时候，你便可以过马路。这些日常生活中的假定对于我们的正常生活是很有帮助的，类似规则或者大家都遵守的潜在契约。

那么什么时候假定是毫无依据的呢？

whenever you take too much for granted – that is, more than is justified by your experience or the particular circumstance.

那么最常见的unwanted assumptions有哪些？

• people’s senses are always trustworthy.[beliefs and desires can distort perception, causing people to see and hear selectively or inaccurately]
• If an idea is widely reported, it must be true.
• having reasons proves that we have reasoned logically.
• familiar ideas are more valid than unfamiliar ones.
• If one event follows another in time, it must have been caused by the other.
• Every event or phenomenon has a single cause.
• the majority view is the correct view.
• the way things are is the only way they should be.
• change is always for the better.
• Appearances are trustworthy.[外在是会误导人的]
• If an idea is in our mind it is our own idea and deserves to be defended.[ideas in our minds, in a large number, are uncritically absorbed from other people. 大部分的我们不过是他人思想的载体，传声器]
• The stronger our conviction about an idea, the more valid the idea.
• if we find an error in someone’s argument, we have disproved the argument.[an argument can contain minor flaws yet be sound]

策略：

Remember that assumptions are usually implied rather than expressed directly, much like the hidden premises in arguments. to identify them, develop the habit of reading(and listening) between the lines for ideas that are unexpressed but nevertheless clearly implied. once you have identified an assumption, evaluate it and decide whether it is warranted.

这里与第八章的argument呼应了，be alert to hidden premises.

The either/or outlook

either/or outlook 是什么？

the either/or outlook is the expectation that the only reasonable view of any issue is either total affirmation or total rejection.

[认为看待一个问题唯一合理的观点要么是完全肯定，要么是完全否定。]

what’s the problem with it?

it rejects the very real possibility that the most reasonable view may be both/and – in other words, a less extreme view.

策略：

whenever you are examining an issue and find yourself considering only two alternatives, ask yourself whether additional alternatives exist and, if they do, give them a fair hearing.

Mindless conformity

盲目遵守或从众。

conformity is behaving as others do.[别人做什么，你便做什么。]

One kind of conformity like looking both ways before crossing the street, make life easier and safer. The imitation of good role models is also a positive conformity. this kind of conformity helps us develop our capacities and become better individuals.

[有些从众是有益的，比如遵守交通规则，过马路前要看看两边，向好的榜样学习等]

但是，mindless conformity, 盲目的遵守，是毫无依据，甚至有时候是不合理的。

it consists of following others’ example because we are too lazy or fearful to think for ourselves.

广告主们最鼓励这种盲从了。

策略：

The secret to avoiding mindless conformity is to resist whatever pleading, teasing, and prodding others exert to make you think and speak and act as they do. Instead of succumbing, ask yourself what is reasonable and right and follow that path, regardless of whether that places you in the majority or the minority.

absolutisim

定义：

Absolutism is the belief that there must be rules but no exceptions.

弊端：

Absolutists tend to be impatient in their thinking and therefore susceptible to oversimplification and hasty conclusions. and once a rule is established, absolutists refuse to allow exceptions.

策略：

when you begin to examine any issue, even one that you have thought about before, commit yourself to accepting the truth as you find it rather than demanding that it be neat and simple.

relativism

定义：

relativism is the polar opposite of absolutism. the relativist believes that the existence of exceptions proves there can be no rules.

弊端：

The belief that truth is created rather than discovered.

Relativism also holds that morality is subjective rather than objective – in other words, that moral rules are binding only on those who accept them.

策略：

remind yourself from time to time that some ideas, and some standards of conduct, are better than others and that the challenge of critical thinking is to discover the best ones.

bias for or against change

the only reasonable attitude toward for or against change is “it depends on what the change is”.

Now bias for change is more common than it used to be ,no doubt because we live in an age of unprecedented change, especially in technology, because many changes are beneficial, we may make the mistake of believing that all are.

but bias against change is still more prevalent than bias for change.【相比拥抱变化，拒绝变化更为普遍】

Why?

• the force of familiarity

  Most of us prefer ideas that we know and feel comfort with.

• our “mine is better” perspective

  New ideas challenges our sense of security, so we tend to resist them.

这里举了好几个例子说明为什么很多人都遵守着传统的习惯，甚至都不曾质疑过，比较常见的一个例子是这样的，女孩的妈妈告诉她，不要把帽子放在桌上或者把外套放在床上，她便遵守了很多年，后来她长大了，成为了母亲，便教育她的孩子也要如此，小女儿便问为什么要怎样，这时她才意识到她从来没有好奇过，更没有问过原因，随后她问已经八十有余的老母亲，才知道原来在她母亲小的时候，一些邻居的小孩感染了虱子，所以她的妈妈才要求她不要把帽子放在桌上或者把外套放在床上。【这个例子有很多延伸的版本，我记得有一个切火腿的例子，基本是同一个套路】

策略：

monitor your reaction to new ideas.

don’t be surprised if you strongly favor or oppose an idea the first time you encounter it. However, refuse to endorse your first impression uncritically. Instead, suspend judgment until you have examined the idea carefully. If the idea proved insightful and well substantiated, accept it regardless of its oldness or newness, if it is flawed, reject it.

chapter 10 errors of procedure

chapter 9, examined errors of perspective, flawed outlooks that create significant obstacles to critical thinking even before we address any issue.

this chapter, examine the kinds of errors that occur in the process of addressing specific issues.

【个人感觉chapter 9中列出那些常犯的错误，基本发现在investigation之前。而这一章则举出那些在investigation之后，人们容易犯的错误】

以下，详细列出6种在下定论过程中易犯的错误。

Biased consideration of evidence

【对于evidence，做偏向性的处理。】

两种形式：

• seeking only evidence that confirms your bias[第一阶段investigation]
• when evidence is presented to you that challenges your bias and you choose an interpretation that favors your bias, even when other interpretations are more reasonable.[第二阶段interpretation]

弊端：

这里举了一个例子说明为什么偏向性的处理论据，会影响我们的judgement。例子是调查一个社会现象：为什么非裔美国人受到犯罪，高失业率，受教育程度低的困扰？调查者在调查中带有坚定的信念，认为贫穷和种族歧视是引发问题的原因。随后作者列出了诸多不一样的观点，包括贫穷与犯罪并没有直接的关系，黑人社区中对于教育以及心智发展的忽视等， 认为一个完全忽视这些异见的调查分析是不公正，且不负责任的。

The worst aspect of bias is that if often occurs innocently, without one’s awareness. And not just among students, even professional scholars can commit this error.

策略：

begin your investigation by seeking out individuals whose views oppose your bias and then go on to those that support it. Also, choose the most reasonable interpreatation, regardless of whether it flatters your bias.

[这个适用于所有的errors吧？]

double standard

【双向标准，对两种ideas采取不同的判定标准。】

double standard consists of using one standard of judgement for our ideas and ideas compatible with our own and an entirely different – much more demanding – standard for ideas that disagree with ours.

【宽以待人，严于利己，真的是太对了】

这个弊端太明显了。

策略：

decide in advance what judgement criteria you will use and apply those criteria consistently, even if the data in question do not support your view.

hasty conclusion

Hasty conclusion is a premature judgment — that is, a judgment made without sufficient evidence.

Many people are in the habit of accepting the first judgment that comes to mind, never bothering to inquire

Whether a different judgment might be as reasonable or perhaps even more so.【思考第一反应的合理性】

【草率定论】

这里回到了第一章提到的那个例子，认为智力是天生的，不可改变的，随后基于这个观点，做出了一系列不合理的举措。

策略：

identify all possible conclusions before you select any one. Then decide whether you have sufficient evidence to support any of those conclusions and, if so, which conclusion that is.

Remember that there is no shame in postponing judgment until you obtain additional evidence.

overgeneralization and stereotyping

【翻译下：以偏概全及坚持以偏概全不动摇，哈哈】

generalizing is the mental activity by which we draw broad conclusions from particular experiences.

【从特定的经验中提取出一般性的结论】

generalizing is not only natural but indispensable to learning. 【我们不可能看到一个事物的所有实例，比如并不需要看到这世上所有的狗才能了解狗，在有限的时空中，认识可见的实例即可，当样本量开始增大时，随时准备好修正原始的认知。】

generalizing挺好，但是over了就不好了。

It is easy to overgeneralize – that is, to ascribe to all the members of a group what fits only some members.

好比你认识一个邋遢的韩国人，然后就断定所有的韩国人都邋遢，这显然是不对的。

a stereotype is an overgeneralization that is especially resistant to change.【坚持以偏概全不动摇】

the most common types of stereotypes are ethnic and religious. 【最常见的模式化观念是种族和宗教】

弊端：

both overgeneralizations and stereotypes hinder critical thinking because they prevent us from seeing the differences among people within groups.

策略：

resist the urge to force individual people, places, or things into rigid categories. In forming generalizations, keep in mind that the more limited your experience, the more modest you should make your assertion.

oversimplification

simplification is not only useful but essential, particularly at a time like the present, when knowledge is expanding so rapidly.【大道至简】

oversimplification, on the other hand, goes beyond making complex ideas easier to grasp; it twists and distorts the ideas. Instead of informing people, oversimplification misleads them.【过于简单以至于都会误导他人】

Oversimplification often occurs in matters about which people have strong feelings.

策略：

Be alert for oversimplification in what you read and hear, and avoid it in your own thinking and expression.

post hoc fallacy

这个出自拉丁术语的缩写，意思是” after this, therefore beacuse of this.”,直白点就是说，B发生在A之后，那么A就是B发生的原因。这个在chapter 3的understand cause and effect中提到过。事情发生的先后跟因果可能没有任何关系。

The error in this thinking is the failure to realize that mere order and closeness in time does not prove a cause-and-effect relationship.

这种想法是大多数迷信的基础，认为事情的发生都有关联，比如镜子破了，今天有霉运了。

you should be careful to avoid the post hoc error ——withhold judgment until you have evaluated all possible explanations, including coincidence.

chapter 11 errors of expression

这里需要明白的一点是，在表达中出现的errors也是thinking errors的一种，因为他们源于你的思想。

they originate in the mind, more or less consciously.

具体有哪些？

• contradiction【自相矛盾】
• arguing in a circle
• meaningless statement
• mistaken authority
• false analogy
• irrational appeal【不合理的诉求】

来看看每种error的定义，弊端以及对应策略。

contradiction

The principle of contradiction is one of the fundamental principles of logic.

no statement can be both true and false at the same time in the same way.

the best way to see its correctness is to try to construct a statement that disproves it.

举了三个例子，这里提一下第一个：

argument：O. J. Simpson 谋杀了Nicole Brown Simpson

这里，O. J. Simpson要么谋杀了Nicole Brown Simpson ，要么就没有谋杀，二选一，但是如果O. J. Simpson雇佣别人去谋杀了Nicole Brown Simpson呢？这样的话，O. J. Simpson到底是不是谋杀了Nicole Brown Simpson？ 这时候，你就不可以说，他既谋杀了Nicole Brown Simpson又没有谋杀Nicole Brown Simpson。他不是真正的实施者，算是没有谋杀？但是他是推动者，要为Nicole Brown Simpson的死负责，算是谋杀？更合理的表达可以是：he would have murdered her in the sense of being responsible for the act but not in the sense of having carried it out.

在你的表达中，警惕出现自相矛盾。

如何避免？

monitor what you say and write.

The moment you detect any inconsistency, examine it carefully. Decide whether it is explainable or whether it constitutes a contradiction. If it proves to be a contradiction, reexamine the issue and take a view that is both consistent and reasonable.

arguing in a circle

用不同的方式去重复同一个论点，而不是给出对应的论据。

A person arguing in a circle attempts to prove a statement by repeating it in a different form.

如何避免？

To detect circularity in your writing or speaking, it is not enough to read and nod in agreement with yourself. You must check to be sure the evidence you offer in support of your view is not merely a restatement of the view in different words.

meaningless statement

meaningless statement更像是毫无意义的废话，严肃的学术论文中，自然应该避免，但是一般的文学作品中，meaningless statement还是很多的。

in the course of presenting ideas, people often find it useful or necessary to present the reasons that underlie their thoughts and actions. A meaningless explanation is one in which the reasons make no sense.

如何避免？

to detect meaningless statements in your writing, look at what you have said as critically as you look at what other people say. Ask, Am I really making sense?

mistaken authority

own authority to someone who does not possess it.

比如问一个扮演医生的演员，这个角色让他学到了什么是可以的，但是问他那一类需要专业性知识才能解答的问题，比如医患关系，就不合理的。

如何避免？

to avoid the error of mistaken authority, check to be sure that all the sources you cite as authorities possess expertise in the particular subject you are writing about.

false analogy

错误的类比。

an analogy is an attempt to explain something relatively unfamiliar by referring to something different but more familiar, saying in effect, “this is like that.”

好的类比，可以让人瞬间解惑，而不恰当的类比会误导他人。

An analogy is acceptable as long as the similarities claimed are real. 【不过，好的类比需要对两种事物都有很好的理解，正确地get到两者的相似处，还挺难的，可遇不可求】

比如典型的一个错误类比：Animals, which move, have limbs and muscles, the earth has no limbs and muscles, hence it doesn’t move.

如何避免？

Always test your analogies to be sure that the similarities they claim are real and reasonable and that no important dissimilariteies exist.

irrational appeal

不合理的诉求。

An irrational appeal encourages people to accept ideas for some reason rather than reasonableness.

Such an appeal says, in effect, “there’s no need to think critically about this idea or compare it with alternative ideas – just accept it.”

不合理的诉求中最常见的类型，有如下这六种：

• irrational appeal to emotion

  irrational appeal to emotion uses feelings as a substitute for thought。

  用feelings来代替思考。比如一个律师为他的客户辩护，描述他的种种优良品质，对父母的爱，对动物的爱，以引起陪审团的同情，忘记那些指控他的证据，那他犯了什么罪？他谋害了他的父母，竟然还以他现在是个孤儿为由来请求法庭怜悯。【我擦嘞。】

  更合理的方式：

  A rational appeal to emotion not only stimulates feelings but also demonstrates their appropriateness to the ideas being presented.

• irrational appeal to tradition

  an irrational appeal to tradition urges maintaining the tradition merely because we’ve always done so. [因为传统一直如此，所以我们也应该这样。如果真的是这样，人类大概是不会进步了，因为要保持传统啊，哪一次革命性的进步不是对传统的革新呢？]

  更合理的方式：

  An appeal to trandition must not only tell people how old and revered the tradition is but also how that it still deserves our endorsement.

  合理的诉求会告诉你为什么这样的传统依然值得我们认可。

• irrational appeal to moderation

  An irrational appeal to moderation is offered on the erroneous presumption that moderation is always preferable.

  基于错误的假设，认为适度总是可取的。【过于看重适度了。】

  书中举了奴隶制度为例，有些人提出可以适度的废止奴隶制，也就是允许一部分人有权决定是否拥有奴隶。他们忘了，奴隶也是同样的人，他们有权利决定是不是愿意被奴隶主所拥有。

  适度不是通用的。

  更合理的方式：

  includes an explanation of why the more moderate idea or action is preferable to less moderate alternatives.

• irrational appeal to authority

  此处authority 可以是人，书，或者文档，机构。

  An irrational appeal to authority says, “Here is what one or more authorities say — accept it unquestonly”。对于权威所说的，没有任何质疑，全盘接受，基于part1的洗脑，这种诉求明显不能接受。

  更合理的方式：

  Says “Here is what one or more authorities say,” and proceeds to show why that view should be accepted.

• irrational appeal to common belief

  这种诉求犯了unwarranted assumption，因为大多数人都相信，所以相信吧。

  An irrational appeal to common belief says, “Believe this because most people believe it”.

  更合理的方式：

  says “Moset people believe this”, and goes on to show the reasonableness of the belief.

• irrational appeal to tolerance

  An irrational appeal to tolerance says, “Because tolerance is good in general, it is the right response to every situation, including this one.”

  我觉得这里连前提都是有问题的，tolerance并不是回应所有情况的合理反应。

  更合理的方式：

  explans why tolerance is appropriate in the particular situation in question.

总而言之，判断一个诉求合理与否，看它是否给到了相应的解释。

the best way to distinguish between rational and irrational appeals is to ask whether the appeal is accompanied by an explanation of why you should accept it.

chapter 12 errors of reaction

so far, 7 errors of perspective, 6 errors of procedure, 6 errors of expression .

何时发生？

errors of reaction occur after we have expressed our ideas and others have criticizes or changenged them.

为什么会发生？

whatever does not flatter us or our point of view reflects our urge to save face and preserve our self-image.

each of us has a self-image, generally a favorable one.

some people manage to resist the tempation to save face, but most of us fall victim to it from time to time.

For still others, it is neither the particular aspect of the image nor the role involved that triggers the face-saving reaction. it is the people who are observing.

总的来说，主要是face-saving.

有哪些类型？

• automatic rejection
• changing the subject
• shifting the burden of proof
• Straw man [稻草人]
• attacking the critic

来逐个了解下。

automatic rejection

可以粗暴地类比成：非我族类，我不听，我不听，我不听。

开启自动反对模式。

automatic rejection is to reject criticism without giving it a fair hearing.

合理的做法应该是：evaluate the challenges and make an honest determination of its worth.

这里作者举了一个他朋友和他同事的例子，说明automatic rejection的诱惑不易抗拒。

如何避免？

A good way to lessen that temptation is to put some emotional distance between your ideas and your ego. Think of them as possessions that you can keep or discard rather than as extensions of yourself. This will make your less defensive about them.

changing the subject

转移话题。

Changing the subject consists of abruptly turning a discussion in a different direction.

转移话题并不总是会犯错。

changing the subject is an error only when the original issue is appropriate and the shift is used deceptively.

【我觉得如果是为了避免自己陷入窘困或者被质疑，感觉到自己的观点被挑战，故意避而不谈，转移话题，这种才算是errors】

如何避免？

Intentionally changing the subject frustrates the purpose of discussion. To avoid this error, face difficult questions head-on.

if you know the answer, state it. if the issue is too complex to permit a certain answer, state what you believe to be probable and explain your reasoning. if you lack sufficient knowledge to speak of probabilities, say so. No reasonable person will think less of you for candidly admitting ignorance.

shifting the burden of proof

The errors of shifting the burden of proof consists of demanding that others disprove our assertions.

当别人要求你提供论据时，你的反应是，你能给出反对我论点的论据？如果不能，就接受。这就是转移论据的责任。

如何避免？

you wll be less likely to shift the burden of proof if you learn to expect your ideas to be questioned and criticized and prepare to support them before you express them.

straw man

稻草人原意是指用稻草作出模型，穿着人类的衣服，放在田间吓走小鸟的那种。这里的straw man具体指：

put false words in someone else’s mouth and then expose their falsity, conveniently forgetting that the other person never said them.

【有些强加给他人错误的观点，然后指出他人的错误的味道，这戏有点多啊。】

如何避免？

be scrupulously accurate in quoting or paraphrasing other people’s words.

attacking the critic

这个其实老祖宗就说过，对事不对人。

别人一反对你的观点，你就开始攻击别人，太幼稚了。

Attracking the critic is the attempt to discredit an idea or argument by disparaging the person who expressed it.

如何避免？

understand that ideas and people are not synonymous. it is unreasonable to substitute speculations or judgments about people themselves for judgments of their ideas.

Chapter 13 The errors in combination

这一章是对前面几章所列出的errors的梳理，前面都是单独去介绍每一个error，但是现实中，往往是多个error同时出现。这一章先整理了前面列出来的那些坑，随后举了3个实例来学以致用，最后对比了几个容易混淆的概念。

Part 2 中提到的所有error及对应的处理方法：

The most fundamental critical thinking error is “Mine is better” thinking. in which we assume that our ideas must be superior to other people’s simple because they are our ideas. In reality, of course, our ideas are as likely to be mistaken as anyone else’s. To overcome “mine is better” thinking, we must be as critical of our own ideas as we are of other people’s.

sample combinations of errors

这里举了3个样例，我们看看第二个：

Sam 13岁的时候，并不真的想抽烟，但是他的朋友怂恿他去做，然后他做的还不错，渐渐地，由偶尔抽烟到每天一包，抽烟的成本上升，他开始从父母那里偷钱买烟，对此他的解释是要么拿父母的钱然后抽烟，要么就不抽烟，但是他并不打算不抽烟，所以没有办法，只能从父母那里拿钱。现在Sam40岁，结婚有孩子，依然抽烟，他患上了哮喘，但是他自己将这归于过敏，每一个新的外科医生都会告诉他吸烟的危害，他反驳，“他们无法证明吸烟会导致任何疾病，所以应该由个人自己决定是否会受到伤害”。当烟草公司被指控加入尼古丁并禁止报道不利的检测结果时，Sam为他们辩护，这些人这么富有，他们没有理由伤害数以万计的同类。

这里，Sam踩了哪些坑？第一个就是mindless conformity，在朋友的怂恿下，不质疑地接受了抽烟，然后是either / or outlook， 认为只有两种方法去解决抽烟带来的成本问题，要么拿父母的钱继续抽，要么不抽烟，其实可以有第三种，Sam自己去做兼职赚钱抽烟，后面他自己换上哮喘，依然固执地将这个归于过敏，这是face-saving，他对外科医生有关吸烟有害的回应，犯了unwarranted assumption的错误，认为每个人都了解吸烟的方方面面，有权利去决定是否被伤害，而实际上很多人可能并不真的了解吸烟的危害。对于烟草公司的问题，他也是毫无根据地假定富有的人是不会做为了赚钱做坏事的。

A sensible view of terminology

三个容易混淆的errors。

Knowing the right terminology is advantageous, but more important is recognizing where reasoning has gone wary and being able to explain the error in terms of the issue involved.

长文慎入。

Chapter 1: who are you?

开篇提到了两个问题：

• who are you? not just your name, but all the informations about you.
• how did you get the way you are?

重点在于，你是如何变成现在的模样的？是你自己选择的结果吗？[I am this way because I choose to be]

相对于个体，个体所处在的世界对其的影响远比我们想象的要大。

主要的影响因素：

• time and place

  Living in a differect age or culture would make you a different person.

  你的成长环境，生于何时何地，限制了你的经验以及你的思考模式。

  you are the product of your environment。

• idea

  当一个想法被表达出来的时候，与之紧密相关的想法也会不可避免地传达出来。

  比如”if you believe in yourself, you can succeed in anything”, closely related idea” nothing else, but belief – neither talent nor hard work – is necessary for success.” 很明显，这个related idea会把人带入误区。如果换成：”Belief in oneself is an important element in success.” 就合理多了。

  书中举了一个例子，说的是一个多世纪前，心理学上很流行的一个观点：“Intelligence is genetically determined and cannot be increased.”智力是天生的，无法提升的. 甚至有立法者基于这个观点，修订了移民法，来减少因为移民而带来的本国平均智力水平的下降……

  有些时候，你的行为可能在有意识或者无意识中受到这些idea或者related idea的影响。

  The innumerable ideas you have encountered will affect your beliefs and behavior in similar ways––sometimes slightly, at other times profoundly. And this can happen even if you have not consciously embraced the ideas.

• Mass Culture

  这里的mass culture，指的是broadcast media, newspapers, magazines, Internet and popular music, etc.

  生在这个时代，我们的很多时间都浸泡在mass culture中。

  Mass culture大部分都以视频或者音频的形式出现，通常采用高频率的变换来让受众保持兴趣。这样带来的影响是prevented many people from developing a mature attention span. 即不能持续专注在一件事情上。对于students或workers而言，相比mass culture，classes 和work就显得相对无聊了。

  「个人感觉，mass culture让我们陷入了对即时反馈的上瘾模式，希望自己的需求立马可以得到满足，很容易缺乏耐心，而学习和工作都不是立马就出现效果的，反而需要你花费大量的时间和精力，最后的结果可能还不如意，但是后者必然是值得你去投入的地方，当然我觉得mass culture没什么不好，很多好的东西都来自于mass culture，它们更多的像是工具，你怎么用才是关键，好比同样是刀，医生拿来救人，匪徒拿来伤人。呃，又跑偏了，这里讨论的是mass culture对个体的影响，不是它好不好。」

  这部分还提到了政客们如何利用media及言论来影响群众。

• 操纵的“科学”

  提到的两个例子：巴甫洛夫的狗（Pavlov‘s dog）, 约翰·华生的幼儿研究实验（John Watson：sychological Care of Infant and Child）。

  这一部分，有一个很赞的类比：

  把人的大脑比做DVD播放器，每一个想法就好像是打包好的DVD。【这本书如果很多年后小孩们再读，估计都要Google下什么是DVD了吧？】

• 心理学的影响

  我们所处时代的心理学及社会学理论对我们信念的影响。这部分对比了父辈和我们这一代人对于自尊，自律等品性的不同态度，这种对比结果放在现在也适用。我觉得这部分很特别的一个例子在于self-image，个体对于自我的定义很大程度上会影响其行为，促使他/她朝着这个self-image的方向发展，有意识或者无意识地去强化这个image，然后循环这个模式。书中提到了马斯洛的需求理论，最顶层的自我实现self-actualization。与其同时代的Viktor Frankl提出了另一个理论，人类需求中高于self-actualization的self-transcendence， 超越个体，忘记个体，专注外在，舍己为人，有点佛家的味道。「这部分不是很懂」

  第9版与第6版相比，有一部分remove了，说的是儿童时代的生活对个体的影响，会让个体形成四个观点，补上。

  • I‘m not OK — You’re OK【出现在出生及2～3岁的孩童期】
  • I’m not OK — You’re not OK 【缺乏来自父母的鼓励，不能接受他人的肯定】
  • I’m OK — You’re not OK【受到父母的责骂与虐待，导致逃离心态，更多的是通过自我鼓励与自我安慰，而拒绝他人】
  • I’m OK — You’re OK【健康心态】

成为一个独立个体：

需要接受的一个观点：Individuality is not something we are born with but rather something acquired – or, earned. Individuality begins in the realization that it is impossible to escape being influenced by other people and by circumstance.

书中给到了方法论，4点：

• Treat your first reaction to any person, issue or situation as tentative
• Decide why you reacted as you did
• Think of other possible reactions you might have had to the person, issue or situation
• Ask yourself whether one of the other reactions is more appropriate than your first reaction

用我的理解简单复述下：

• 不要急于表达你的第一反应，除非你已经检验过，否则拒绝接受第一反应；
• 想想为什么你会有这样的反应，思考下是否是借鉴于他人，父母或者朋友，或者某个电视里的角色，书中的角色，又或者是什么样的特殊经历让你有这样的反应；
• 想一想，针对其他的人，事，情境，你可能会有的其他反应是什么；
• 问问自己，这些其他反应中，是否有比第一反应更适合的反应？当你寻找答案时，抵制你所处的环境对你的影响「或者说让自己成为一个白板，这个还是很有难度的。」

Practice。

本章最后的Application部分，添加了note，提到了recording的必要性，深以为然。记录下你的想法，你的疑惑，你的观察，定期进行回顾和思考。「吾日三省吾身，真的做起来，是很不容易的。可是所有值得去做的事情，哪一样不需要花费大量的时间和精力？」

Chapter 2: What is critical thinking?

开篇给到了一个场景，小男孩被老师教育着要集中精神思考，随后引出了thinking 的定义。

Mind, Brain, or Both?

在定义thinking之前，给到了一个问题：思考是一种脑力活动，那么思考是属于mind还是brain，抑或两者兼而有之？brain是一个实体，可触可感，但是mind，则是归属于形而上学之类。后面并没有就brain和mind给到结论，只是说明了下，本书是以mind is active, a vehicle by which we take the initiative and exercise our free will为基础。「从这里也可以看出来，作者很看重个人的主动能动性，思考能力是可以通过练习提高的。」

Critical Thinking Defined

• thinking VS feeling

  Feeling is a subjective response that reflects emotion, sentiment, or desire; it generally occurs spontaneously rather than through a conscious mental act.

  Feelings 更多的是对事物的主观反映，折射着人的情感，情绪，欲望，自发产生而不经过有意识的智力活动。好比人饿了就会想要吃饭，渴了就想要喝水，并不会在吃饭喝水之前，做一番思考，因为1，2，3等等原因，所以我决定吃饭喝水。【那种本能的反应，我猜是源自于基因，物种为了在大自然中生存下来而作出的选择，我们的祖先在远古时代每次遇到肚饿的情况，都是去填饱肚子，这个模式不断被调用，就好比经过了无数次的训练，这样经过了几万年的进化，这种对于肚饿的反应就内化成了我们的本能？我猜的】

  feeling并不能替代thinking，因为它并不可靠。书中举了足球运动员Zinedine Zidane在自己职业生涯的最后一个季度不合理地作出了犯规行为的例子，以证明论点，即feeling is notoriously unreliable， 甚至we often feel like doing things that will harm us.

  相比feeling，thinking is a conscious mental process performed to solve a problem, make a decision, or gain understanding。

  feeling与thinking 的关系：

  feelings need to be tested before being trusted, and thinking is the most reasonable and reliable way to test them.

• three categories of thinking

  reflective, creative, and critical.

  this book is on critical thinking.

  The essence of critical thinking is evaluation.

Characteristics of Critical thinkers

misconceptions on critical thinking：

• being able to support beliefs with reasons makes one a critical thinker.【yep，but the reasons should be good and sufficient】
• critical thinkers never imitate others in thought or action。【critical thinking means make sound decisions regardless of how common or uncommon those decision are】「做出与他人同样的决定，并不能认定就是效仿他人吧？同样，critical thinker也不是说一定就是那种特立独行的人」
• critical thinking is synonymous with having a lot of right answers in one’s head.【critical thinking involves the process of finding answers when they are not so readily available, and I think this process is the most important part】
• critical thinking cannot be learned【critical thinking is a matter of habit】

Characteristics of Critical thinkers:

• Skill in asking appropriate questions
• control of one’s mental activities

John Dewey 将这种mental discipline 与freedom挂钩。「思想领域的自由？」

他认为没有这种能力的人只是环境或者冲动情绪的奴隶，不是自由人。「好吧，我不是自由人，我容易冲动」

The Role of Intuition【直觉的角色？】

这部分列举了很多例子，包括日常生活中的场景以及为人熟知的几个科学家的故事，德国化学家Kekule在梦中悟出苯分子的环状结构，作家歌德因一位好友自杀而悟出如何组织大量素材的创作经历，英国诗人塞缪尔泰勒从梦中醒来后脑中浮现200～300行新诗。表达一个论点：

Breakthrough ideas favor trained, active minds。

或者说，sometimes, intuition is the result of thinking.

但是大部分的时候，没有经过刻意训练的直觉，很可能是错的，而且我们会有意识地去遗忘这种错误，从而继续相信直觉。

[有的时候，我们相信专家或者大牛的话，也许有一部分的原因是因为我们相信他们的结论是经过思考后得出的，即使不是经过思考的，单纯凭直觉，也会觉得比一般人靠谱，为什么？认为他们的直觉不是突然从石头里蹦出来的，而是长期在专业里浸泡，思考的产物]

Basic Activities in Critical Thinking

Investigation, interpretation, judgment.

【我感觉这部分算是最核心的了，后面基本都是围绕着这三个部分来的，Part2中列出的思考前或过程中常常犯的错误，不是少了某个环节，就是某个环节没有做到位，不符合要求，而Part3给到的方法论，也是如何做好这三个环节，把一个结论抽丝剥茧，拆分拆分再组装】

Critical Thinking and Writing

这一部分写的真好，都想全文摘抄……

通过writing去发现ideas，不仅仅是排列出你已有的ideas，同时还可能会 stimulating the flow of new ideas。just like an idea machines.文中对idea的产生是这样描述的：

If ideas come slowly, be patient.

If they come suddenly, in a rush, don’t try to slow down the process and develop any one of them, simple jot them all down.

Critical Thinking and Discussion

这部分很形象地举了一个电视上常见的讨论show，当然那讨论的目的早已经脱离了问题本身，不是以解决问题或者加深对问题的理解从而推动问题解决为目的的。文中给了方法论，来避免无效甚至有害的讨论。

• Whenever possible, prepare in advance【提前准备，其实并不会花太多的时间】
• Set reasonable expectations【Expect to have your ideas questioned, and be cheerful and gracious in responding】
• Leave egotism and personal agendas at the door.【有效讨论的前提，有礼貌，尊重对方】
• Contribute but don’t dominate.【善谈者克制泛滥的表达欲，寡言者甩掉顾虑，多多参与】
• Avoid distracting speech mannersms. 【是讨论，不是演讲】
• Listen actively【专心听别人所说的，尝试着进入他人的思维框架中，一旦思绪乱飘，拽回来】
• Judge ideas responsibly【客观判断ideas，不管是自己的还是别人的】
• Resist the urge to shout or interrupt【我有打断别人思路，提问题的毛病，这点要注意改正】

Avoiding Plagiarism【避免抄袭】

这部分强调了idea是一种智力财产。举了好一些例子来否定抄袭行为，并将某些场合下的抄袭定义为一种犯罪行为。

「想到了一句古语：千古文章一大抄。我觉得要看抄袭是为了什么，如果是为了一己谋私，而损害了他人的利益，那确实是不应该的。可是如果说完全的避免抄袭，否定一切的抄袭行为，我觉得就过度了。因为很多时候，你甚至都不知道自己在抄袭，以为自己是原创，后面才发现原来很多东西古人或者同时代的其他人早已经想到了。都是人类的种，基因里总是会有一些重合度颇高的东西。GitHub上那么多的repos，有多少是真正完全原创的呢？或多或少都会带有前人的影子，但这并不妨碍它本身的价值。我想我会有这样的想法，或者说我与作者之间产生的分歧在于我们对抄袭的定义不同吧。因为我将所有言他人所言的都归为抄袭，其实不是，抄袭意味着窃取他人的成果。」

书中对于因为误解了概念而无意识产生的抄袭行为，提到了如何避免的方法：

• When you are researching a topic, keep your sources’ ideas separate from your own.
• As you read each source, note the ideas you want to refer to your writing.
• When you compose your paper, work borrowed ideas and words into your own writing by judicious use of quoting and paraphrasing. In addition, give credit to the various authors.

书中还给到了一个范例，操作起来其实很费时，不过严肃性的研究论文确实应该这样。

最后的Application部分，更多的是自问自答型，并给到了一个如何审视自己思考过程的方法。

Chapter 3 : What is Truth?

开篇提到了what is truth? the existence of truth, could it possible for us to know truth? etc.

最后的焦点集中在 truth is relative and subjective, Everyone creates his or her own truth.

接下来是作者从多个角度解释为何”everyone makes his or her own truth”是一个很可笑的观点，整个论证过程，证据充足，解释合理，堪称精彩，非常值得学习。

Where does it all begin

如果仅仅关注成年时代，Everyone creates his or her own truth 可能听起来比较合理。【书中没说为什么比较合理，也许是成年后，人大多拥有了自己独立的思考能力，而不再盲从，不会太轻易受到外界的影响】

强调童年时代我们所受到的教育对于成年后的我们依然发挥着很重要的影响。

we are told about the world before we see it. We image most things before we experience them. And those preconceptions, unless education has made us acutely aware, govern deeply the whole process of perception.

那些我们认为自己独有的观点中，或多或少都带着前人的印记。

To a greater or less extent, what we regard as our unique perspective bears the imprint of other people’s ideas and beliefs.

Imperfect Perception

本节举了几个例子说明为什么感觉是有瑕疵的，不是完美的。

主要原因两个方面：

• it is influenced by our desires, interests and expectations
• a number of factors make us see and hear inaccurately

Imperfect Memory

Our memory often distorts the data.

我们会给自己的记忆添砖加瓦，甚至有意或者无意美化某段记忆。

We forget details, and when later attempting to recall what happened we resort to imagination to fill in the blanks.

When we talk with others about our experience, we trend to make it more simple and interesting than what we really said or did.

Deficient Information

The quality of a belief depends on a considerable extent on the quality of the information that backs it up.Because it’s a big world and reality has many faces, it’s easy for us to be misinformed.

Things keep happening too fast.

Even the Wisest can Err

圣贤也会犯错。

Truth is Discovered, Not created

这部分算是对上面几个部分的小结，同时引领了下一节。

why “everyone creates his or her own truth” is laughable?

• our ideas and belifs are unavoidably infuenced by other people’s, particularly in childhood
• perception and memory are imperfect
• our information can be inaccurate or incomplete
• some people’s thinking skills are woefully meager(粗劣;少) and/or ineffectively used

What is the most reasonable view of truth?

The truth about something is what is so about it - the facts in their exact arrangement and proportions

Our beliefs and assertions are true when they correspond to that reality and false when they do not.

To look for the truth in such matters is to look for the answer that fits the facts, the correct answer.

Truth is discovered, not created.

Much of the confusion about truth arises from complex situations in which the truth is difficult to ascertain or express.

A good way to begin the right journey is to keep the following thought in mind:

I know I have limitations and can easily be mistaken. And surely I’ll never find all the answers I’d like to. But I can observe a little more accurately, weigh things a little more thoroughly, and make up my mind a little more carefully. If I do so, I’ll be a little closer to the truth.

Understanding Cause and Effect

some of the most difficult challenges in discovering truth occur in determining cause and effect relationships.

Four facts need be understood before discovering the truth:

• one event can precede another without causing it.[比如认为今天打碎了镜子所以霉运连连]

  correlation does not prove causation. [相关关系不等同因果关系]

• Not all causation involves force or necessity[认为causation都是物质的，有形的外在力量]

  Causation also occurs in the nonmaterial realities we call human affairs –more specifically, in the processes of emotion and thought.

  The influence(cause) does not force the effect to occur but instead invites, encourages, or inspires it.

• There is a wild card in human affairs – free will [选择态度的自由]

  free will : the capacity to respond in ways that oppose even the strongest influence.

  Possessing free will is no guarantee that we will apply it. In fact, one factor makes such application difficult. That factor is habit. [习惯的力量]

  最难改掉的习惯是那些随着时间而不断增强的习惯。「复利效应」

• Causation is often complex[起因常常很复杂]

  举了很多个例子，比如车子连环追尾，患流感的女性对同舱人的影响，还就欧洲被中东化做了一番精彩的分析。

Cautions：

Keep the following cautions in mind to avoid oversimplification in your analyses.

• Remember that events seldom, if ever, “just happen” [ 万事皆有因果]
• Remember that free will is a powerful causative factor in human affairs[不要忽视free will]
• Be aware that in a chain of events, an effect often becomes a cause.[在一条事物链中，一件事的果，可能会是另一件事的因]
• Be aware that ,in dealing with human affairs, outcomes can be unpredictable.[涉及到人类的活动时，结果可能无法预测]

[分清事实和看法]

Chapter 4 What does it mean to know?

开篇依然是一个场景，以为自己知道embarrass如何拼写的Sally发现自己是错的，而认为自己不知道embarrass如何拼写的Marie却被证明是正确的。

Requirements of knowing

知道不仅仅是说你知道正确的答案，还包含着你知道你知道正确答案这件事，并且懂得如何去表诉你是如何知道正确答案的。「我这蹩脚的翻译」

Knowing involves more than having the right answer, it also involves the realization that you have it.

Knowing usually implies something else, too - the ability to express what is known and how we came to know it.

Testing your own knowledge

这部分，举了8个例子，说明那些我们以为应该是常识的知识，并不一定就是正确的。这些被质疑或者说推翻的常识里面，我个人印象深刻的是有关情绪的发泄，更具体的是，生气时如何处理。

原来的常识: expressing anger has the effect of reducing it and making us feel better.

反方观点：expressing anger makes you angrier, solidifies an angry attitude, and establish a hostile habit.If you keep quiet about momentary irritations and distract yourself with pleasant activity until your fury simmers down, chances are you will feel better, and feel better faster, than if you let yourself go in a shouting match.

如果我没有记错，《少有人走的路》里面对于情绪的发泄也有一段表述，更多的其实是鼓励情绪的发泄。综合两者，情绪要发泄，但是要以合适的方式，至少不能伤害他人。【呃，这不是废话？】

is it possible that what I and other people think we know isn’t really so?

「电影《the big short》开篇贴出来了那句马克吐温的警言：It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so。可怕的不是你不知道，是你不知道你知道的竟然是错的。」

How we come to know

获取知识的两种方式：主动actively或者被动passively

大部分教室里传授的知识以及通过电视新闻，报刊获取的信息，与身边的人进行的信息交流，都属于后者，passively，被动的获取知识让我们倾向于全盘接受，不带任何批判。

主动的获取包括，类似科学实验那样，自己亲身体验，对于提出的想法进行检测，分析并考虑所有的可能性解读，最终得出符合逻辑的结论。【按这个标准，我感觉自己这些年白活了】

这部分举了一个很常见的场景：一条信息如何在一传十，十传百中，逐渐面目全非的。

Why knowing is difficult

三个主要原因：

• some long unanswered questions continue to resist solution【比如，癌症是由什么引起的？】
• Everyday situations arise for which there are no precedents【某些问题的出现是由于其他一些问题的出现带来的，比如抽烟被认为是诸多致命性疾病的引发因素之一，有关吸烟者与非吸烟者在公共场合的权利争论便开始了，如果没有前者，后者自然也就不存在了。】
• as one generation succeeds another, knowledge is often forgetten or unwisely rejected【这部分，举了很多个例子，不可因为前人犯错，就轻易否认前人的智慧】

A cautionary Tale

这一篇举了一个好玩的例子，十九世纪六十年代，人们在菲律宾的一座小岛上发现了塔萨代Tasaday部落，该部落的人们依然生活在石器时代，大多数的人类学专家和社会学专家在经过实地考察后，认为他们确实是石器时代。而在随后的1986年，一名瑞士的记者发现塔萨代其实是住在房子里，而不是洞穴中，着装也同现代人无疑，他们承认了这一切是由于他们受人压迫，刻意制造出来的假象，但他们也并不完全是伪造，也是处于比较原始的生活状态，只是不算是完全的石器时代了。【这部分去Google了下，当八卦来看，涨知识了……】

Is Faith a form of knowledge?

宗教信仰，意味着相信一些不能被证明的事情。

【一件事并不因为你相信，或者很多人相信，它就是正确的，相不相信跟是不是事实没有必要的联系】

Obstacles to knowledge

阻碍knowledge的两种习惯：

• assuming

  Assuming is taking something for granted.

  assuming is generally an unconscious activity, we are often unaware of our assumptions and their influence on us. the main negative effect of unrecognized assumptions is that they stifle the curiosity that leads to knowledge.

• guessing

  Guessing is offering a judgement on a hunch or taking a chance on an answer without any confidence that it is correct.

  It may denies the importance of evidence.【但是大部分的事实，不都是先猜想，然后验证？】

方法论：what should we do?

Be honest with yourself and others about your ignorance. To admit you don’t know something shows good sense, restraint, and intellectual honesty. These are not weaknesses but strengths.

The admission of ignorance is the essential first step toward knowledge.

you should value firm, bold statements so much that you reserve them for occasions when the evidence permits. Similarly, you should value convictions so highly that you embrace them only when you have sufficient knowledge to do so and that you modify them whenever intellectual honesty requires.

【这一章，感触最深的还是这句：It ain’t what a man doesn’t know that makes him a fool, but what he does know that ain’t so.】

Chapter 5 How good are your opinions?

开篇不再是一个场景，来了一段引言，出自Charles Sylvester de Fort 1931年写的一本小册子，表达的观点是：the earth is flat。

现在看会觉得这样的观点很可笑，但是当你读他写的那段文字，脑子多半会浮现一个义正言辞，捍卫真理的勇者形象，在那个大家不知道地球是圆的的时代，谁说他可笑呢？

由此引出本章重点：how good are your opinions?

这里，作者也很明确划出了critical thinking中所指的opinion:

refers to expressions of judgement rather than expressions of taste.

毕竟有关taste的争论是没有什么意义的。

但是有的时候，很难分辨出他人的观点是有关taste还是judgement的。

「在一个自由的国度，每个人都有拥有自己观点的自由，但是你拥有它并不意味着它就是正确的，如果它是错误的，那么基于这个观点作出的相应行为，很可能会伤人或伤己。」

Being free to hold an opinion and express it does not, of course, guarantee favorable consequence. Acting on our opinions carries even less assurance.

We are free to act on our opinions only as long as, in doing so, we do not harm others.

Opinions can be mistaken

这里举了一个Taster‘s Choice coffee的例子，同样的咖啡被贴上了不同的标签后，人们会尝出完全不同的味道。【谁说成人缺乏想象力？我们想象力丰富着呢。】

既然opinions 可能是错误的，那么citical thinking能做什么？

Ideas are seldom of equal quality. Solutions to problems vary from the pratical to impractical, beliefs from the well founded to the ill founded, arguments from the logical to the illogical, and opinions from the informed to the uninformed. Critical thinking serves to separate the more worthy from the less worthy and, ultimately, to identify the best.

Opinions on moral issues

这是个比较敏感但又很重要的部分，很多时候，我们的观点都会涉及到道德领域。这个时候，如何去判断呢？有的时候，这些涉及道德规范的观点可能是非常主观和完全个人倾向的，如何去判断？或者说以什么样的标准来判定哪种观点更为正确？

文中给了三个最重要的标准：obligations, ideals , consequences.

• obligations：when two or more obligations conflict, the most important one should take precedenct.【更重要的那个优先，可是重要的标准是什么？】
• ideals：when two or more ideals conflict in a given situation, the most important one should take precedenct.【与obligations相同】
• consequences：Actions that achieve beneficial consequences should be preferred over those that do harm. preferred action is the one that achieves the greater good or the less evil.

Even Expert can be wrong

这段举了大量的例子，涉及医学，生物学，心理学，天文，考古等等，来证明作者的观点：even expert can be wrong.

Kinds of Errors

这段还蛮精彩的。

观点容易掉进的四种坑：

• Errors or tendencies to error common among all people by virtue of their being human.【是个人都会犯错……】
• Errors or tendencies to error associated with one’s individual habits of mind or personal attitudes, beliefs, or theories.【与个人思维习惯，个人态度，信仰或者理论有关】
• Errors that come from human communication and the limitations of language.【交流以及语言的局限性，the practice of expressing a thought or feeling inadequately and leading others to form a mistaken impression】
• Errors in the general fashion of an age. 【一个时代的fashion?】

英国哲学家John Locke观察指出，容易犯错的人基本可分为三类：

• Those who seldom reason at all, but think and act as those around them do.
• Those who are determined to let passion rather than reason govern their lives.
• Those who sincerely follow reason, but lack sound, overall good sense, and so do not look at all sides of an issue.

作者加了一个分类，这个类别的人其实才是最难“治愈”的：

• Those who never bother to reexamine an opinion once it has been formed.

Informed versus uninformed opinion

这部分其实跟前面even expert can be wrong相呼应，既然专家也可能是错误的，那么我们为什么还要咨询他们的意见呢？不是浪费时间吗？自然不是，同一件事情，专家与常人的判断相比，更有参考价值，当然，前提的，这个人真的是专家。

【专家相比常人，因为在他们所研究的领域投入的更多的时间和思考，所以他们针对该领域提出的观点，其错误的几率相比其他人要低很多，这也是为什么我们生病了要去看医生，但是不是随便哪个医生都能给你治病，要找对应的专科医生】

咨询专业人士可以有效减少犯错的几率。

By examining the opinions of informed people before making up our minds, we broaden our perspective, see details we might not see by ourselves, consider facts we would otherwise be unaware of, and lessen our chances of error.

这里作者还附送了一条如何写分析报告的建议：

In most responsible expressions of opinion, the statement of opinion takes up only a sentence or two, while the supporting details fill paragraphs, pages, and even entire chapters.

Forming opinions responsibly

这部分算是方法论了。

Forming opinions is natural. Here are some tips that can help you improve the quality of your opinions:

• Understand how opinions are formed

  In practical terms, you cannot help producing opinions about what you see and hear whether or not you take control of the process. when you are not in the control, your mental system operates in the uncritical default mode.

• Resist the temptation to treat your opinions as facts

  The prudent approach is to question any opinion, even a cherished one, the moment evidence arises that suggests it is based on habit, impulse, whim, personal preference, or the influence of fashionable ideas rather than reality.

• Monitor your thoughts to prevent the uncritical default mode from taking charge

  Be not swept off your feet by the vividness of the impression, but say, “Impression, wait for me a little, Let me see what you are and what you represent, let me try test you.”

Gain control of your opinions, rather than having them control you.

Chapter 6 What is Evidence?

大体的内容：什么是证据？证据有哪些类型？如何评价证据？充足的证据需要满足什么条件？

To present evidence is to show others that what we think makes sense.

证据用于表明你的观点不是随口一说，是有依据的， 是用于支撑你论点的那些材料。

为什么很多文章和演讲中包含了大量观点但是却没有提供相应的证据去支持这些观点？

• The human mind is a veritable opinion factory, so most people have an abundance opinions to share.

• people tend to remember their opinions and forget the process by how they got them.

• sometimes there is little or no evidence to remember - in other words, the opinion is based on nothing substantial.【毫无依据的观点也是观点】

  这里举了克林顿的桃花案事件中一些人的看法来佐证。

大部分人都不愿意承认，当我们被询问支撑这个观点的证据是什么时，我们只能举出脆弱的，经不起推敲的证据，我们可能会安慰自己说我们的脑中有足够的证据，但是事实上，更大的可能性是我们脑子只有那些可怜的脆弱的证据。

we may soothe ourselves with the notion that a thick folder of evidence lies misfiled in our minds, but the very real possibility remains that flimsy evidence was all we ever had.

那么critical thinker有什么不同呢？

Critical thinkers are tempted to commit the same self-deception that plagues(纠缠) others. but they have learned the value of resisting that temptation. More important, they have developed the habit of checking the quality and quantity of the evidence before forming an opinion. Also they review their evidence before expressing an opinion.

【critical thinker想得多，花的时间自然也比常人多，估计单身狗居多吧……】

Kinds of Evidence

这部分列出了大部分证据的来源，根据人们对它们的熟悉程度排序，并陈述了每种证据类型的价值，局限性以及要问的适当问题。

具体如下：

• personal experience

  这里的personal experience不仅仅是指你个人的，也包括他人的。

  Value:

  we carry personal experience in our minds, it tends to exert a greater influence than other kinds of evidence. it makes things more authentic and meaningful than what we have merely heard or read.

  limitation:

  Attach greater significance and universality to particular events than they deserve.【容易犯以偏概全的错误，比如认识一个韩国人，就认为我们已经了解了大部分的韩国人】

  appropriate questions to ask:

  • Are the events typical of unique?
  • Are they sufficient in number and kind to support the conclusion?

• unpublished report

  这里的unpublished report指的是stories we hear from other people, 通常指的是那些八卦，道听途说的东西，比如彭于晏是我失散多年的男票。

  Value:

  Offer information.

  limitation:

  difficult to confirm them. 口口相传后，你听到的可能是2.0版本了，即使你听到的是第一版的，也不能保证它是真实的。

  appropriate questions to ask:

  • where did the story originate？
  • how can I confirm that the version I heard is accurate?

• Published report

  已经公开发布的报告，包括博客，正式的学术论文，专业性的报刊等等。

  Value:

  offer information

  limitation:

  主要是非学术类中，事实和观点通常混合在一起，文档可能也是非正式的，残缺不全的，甚至不存在的。

  In nonscholarly works, the documentation may be informal, fragmentary, or, in some cases, nonexistent.

  Facts and opinions are often mingled in contemporary publications, particularly nonscholarly ones.

  appropriate questions to ask:

  • Does the report cite the sources of all important items of information?
  • Does the author have a reputation for careful reporting?
  • Does the publisher or broadcaster have a reputation for reliability?
  • Which statements in the published report constitute evidence, and which should themselves be supported with evidence?

• Eyewitness testimony

  Value:

  commonly considered to be the most reliable kind of evidence. 眼见为实。

  limitation:

  证据的真实性受到外部环境和观察者本人的影响。比如起雾的夜里，看到的也许并不是实物真实的样子,具体而言，有这些可能：

  • The incident may have occurred late on a foggy night and the eyewitness may have been some distance away.
  • The eyewitness may have been tired or under the influence of alcohol or drugs.
  • His or her observation may also have been distorted by preconceptions or expectations.
  • The person’s memory of what occurred may have been confused by subsequent events.

  appropriate questions to ask:

  • What circumstances surrounding the event, including the eyewitness’s state of mind, could have distorted his or her perceptions?
  • what circumstances since the event — for example, the publication of other accounts of the event – could have affected the eyewitness’s recollection?

• Celebrity testimony[名人证词，背书]

  这里指的是网红，歌手，演员，导演，运动员或者行业里面的大牛们， 在一些商业或者正式场合中，为某些产品或者服务宣传背书，当他们被邀请去讨论有关宗教，教育，经济，国际关系等话题时，不要把他们的观点当成事实。

  Value:

  offer information

  limitation:

  they may have been paid to say things or they may just to be impressed.

  appropriate questions to ask:

  • in the case of advertisements or informercials, is the celebrity a paid spokesperson?
  • in the case of talk show comments, does the celebrity offer any support for his or her views? Also, does the host ask for such support?

• Expert opinion

  Value:

  More reliable than most of the varieties of evidence so far.

  it can usually address the crucial question of what is typical and what is not ,which is the advantage over personal experience.

  limitation:

  Knowledge in virtually every field is rapidly expanding.

  some people can’t resist the temptation to think of themselves as experts in everything. so one expert who expertise in a particular issue may write articles on other fields which he doesn’t know well.

  appropriate questions to ask:

  • Does the person have, in addition to credentials in the broad field in question, specific expertise in the particular issue under discussion?
  • whether the expert is paid.
  • Whether other authorities agree or disagree with the expert’s view.

• Experiment

  Value:

  offer information

  limitation:

  two broad types of experiments.

  • laboratory experiment: its artificiality
  • field experiment: the presence of the researchers can influence the subjects and distort the findings.

  appropriate questions to ask:

  • for a laboratory experiment, has it been replicated by other researchers?
  • for a field experiment, have other researchers independently confirmed the findings?

• Statistics

  Joel Best notes： “We think of statistics as facts that we discover, not as numbers we create……statistics do not exist indepently.”

  Value:

  applies to any information that can be quantified.

  limitation:

  Sometimes statistical errors are intentional.

  appropriate questions to ask:

  • What is the source of the statistics?
  • is the source reliable?
  • how old are the data?
  • Have any important factors changed since the data were collected?

• Survey

  Value:

  data obtained from surveys are quantifiable.

  limitation:

  the sampling may be random, systematic or stratified.

  appropriate questions to ask:

  • Was the sample truly representative?
  • were the questions clear and unambiguous?
  • were they objectively phrased rather than slanted?
  • do other surveys corroborate the survey’s findings?

• Formal observation

  two kinds of formal observational studies.

  • detached observation: the observer does not interact with the individuals being studied
  • participant observation: the researcher is involved in the activity being studied.

  Value:

  offer information

  limitation:

  in participant observation, it may distort the individuals’ behavior.

  appropriate questions to ask:

  • is it likely that the presence of the observer distorted the behavior being observed?
  • Was the observation of sufficient duration to permit the conclusions that were drawn?
  • do the conclusions overgeneralize?

• Research review

  This kind of study is undertaken when a considerable body of research has already been done on a subject.

  Value:

  a thorough review of research reveals areas of agreement and disagreement and provides a valuable overview of the current state of knowledge on the subject.

  limitation:

  difficult to evaludate all the reviews

  appropriate questions to ask:

  • do the reviewer’s conclusions seem reasonable given the research covered in the review?
  • has the reviewer omitted any relevant research?

One additional question is applicable to all kinds of evidence:

Is this evidence relevant to the issue under consideration?

Evaluating Evidence

这段列出了哪些情况下，我们可能会带有个人倾向地评估evidence.

bias: we usually be tilted in one direction or the other by our overall philosophy of life, our political or social views, our opinions on related issues, or our attitude toward the people associated with the various views.

signs that show bias in hindering your evaluation of evidence:

• you approach your evaluation wanting one side to be proved right
• you begin your investigation assuming that familiar views will prove correct.
• you look for evidence that supports the side of the issue you favor and ignore evidence that opposes it.
• you rate sources by how favorable they are to your thinking rather than by their reliability and the quality of their research.
• you are nitpickingly critical of evidence for views you oppose and uncritical of evidence for views you favor
• when you encounter evidence that opposes your bias, you begin arguing against it, often before you have completed examining it.

后面提到了evaluating evidence的价值，这段写得真好, 有时候接受事实并不容易，但是总比一直抱有错误的观点要好：

Although you may not be able to eliminate your biases, you can nevertheless identify and control them. and that is all that is necessary.

The purpose of evaluating evidence is to discover the truth, regardless of whether it is pleasant for unpleasant, and the only way to do so is to evalute fairly. Such an evaluation will sometimes require you to conclude that the view you leaned toward(or actually held) is mistaken. When the evidence supports such a conclusion, have the courage to embrace it. Changing your mind is not dishonorable, but mantaining a false view in order to save face is not only foolish but also intellectually dishonest.

What constitutes Sufficient Evidence?

给了一个通用的guidelines 来帮助我们去判定evidence是否是充足的：

• Evidence is sufficient when it permits a judgement to be made with certainty.

  Certainty exists when there is no good reason for doubt, no basis for dispute.

  Certainty is a very difficlut standard to meet, especially in controversial issues, so generally you will be forced to settle for a more modest standard.

• If certainty is unattainable, evidence is sufficient if one view of the issue has been shown to have the force of probability.

• In all other cases, the evidence must be considered insufficient.【to be a critical thinker is really not easy….. 】

Chapter 7 What is Argument?

开篇洗脑，说了下在critical thinking的领域里，argument是什么。

argument的三种含义：

• a quarrel【little relevance to critical thinking】
• the exchange of opinions between two or more people 【relevant to critical thinking】
• the line of reasoning that supports a judgment. 【relevant to the challenge of becoming a critical thinker， 这个是critical thinking中对于argument的定义】

应该说对于critical thinker而言，argument指的是得出具体某个结论的推理过程。

类似数学公式的推导，因为1+1 = 2 所以 2 - 1 = 1 ， argument也是以一个或者多个假定为前提，推导出一个或者多个结论的过程。

相比数学公式的推导，arguments 会更复杂，更难测试其正确性。

The parts of an argument

两个部分：假定premises和结论conclusion。

逻辑学家在评估arguments时的基本原则：

• the premises are either true or false
• the reasoning that links the premises to the conclusion is either valid or invalid.
• correct premises plus valid reasoning equal a sound agrument.
• Either an incorrect premise or invalid reasoning will render an argument unsound.

正确的假定前提加上有效的逻辑推理能得出一个合理的论点，但是有时候，正确的信息也可能会推断出错误的结论。

Inappropriate attitudes toward ideas and the reasoning process can also lead to errors in argument. 想法以及推理过程中不恰当的态度也会导致论点错误。这里举了第一印象为例，不要以第一印象给他人盖棺定论。

Evaluating Arguments

评估论点的方法其实前面已经提到了，很简单：

Decide whether the premises are true or false and whether the reasoning that leads from them to the conclusion is valid.

这段举了好几个例子来演示如何去评估一个论点，这里挑一两个列出来。

这里假定Bruno没有通过测谎仪测试，前提都是正确的，但是最后的结论却是不对的，第一个前提说犯罪的人通常不会通过测谎仪测试，premise中有一个usually ，由此推断出来的结论只能是Bruno可能是有罪的，并不能肯定，也就是可以推出另一个结论，Bruno是无辜的，清白的。即这两条假定不足以推断出这样的结论。

后面提到了对立的两个论点可以帮助我们更有效去推断更正确的结论：

Did you ever have the experience of hearing an argument on some issue, being impressed with it, and then hearing the opposing argument and being even more impressed with that? 【 妥妥地中枪，我同意A，我觉得A的反面也很有道理】It happens often.【文中没有给到处理方法，只是举例说明了这种情况经常发生】

Remember that your evaluation of any argument is likely to be most effective when you are able to hear both sides or at least to consider the criticisms people on each side of the issue make of the other side’s view.

More difficult Arguments

列出了推断论点的过程中，比较难的几种情况以及相应的处理方式：

• When an argument is longer than a paragraph

  summarize it before asking and answering your questions.

  the danger in summarizing, is that you might misrepresent what the person was saying. if you are careful, you can avoid this misstep.

• When you are uncertain which statements are the premises and which is conclusion

  ask yourself exactly what idea the person is trying to get you to accept(that is the conclusion).

  ask what reasons are offered in support of that idea.(those are the premises)

• When an argument contains more than two premises

  ask and answer your questions about each. don’t be daunted if there are many premises.

  after eliminating any irrelevant premises, decide whether the conclusion follows logically from the remaining premises and if it is the only conclusion that does.

  if more than one conclusion follows, decide whether the stated one is the most reasonable conclusion.

• When you are evaluating opposing arguments

  neither of which is persuasive(有说服力)， look for a third alternative.

  【感觉如何分辨两个完全相反的论点，给出最合理的结论还是挺困难的，难点在于如何去判断论点的不足在哪里。】

  这部分举了一个例子，有关在公立学校展示十诫的讨论，涉及到宗教以及道德这块，作者给出的解决方法是列出所有版本的十诫以及renew其他宗教或道德的清单，逐个进行推理。

  accommodating all perspectives is no more offensive than ignoring all and has the additional benefit of emphasizing the importance of moral values.

• When an argument contains hidden premises

  identify them before proceeding with your evaluation.

  辨别出那些隐藏的假定前提对于推理出结论来说，很重要。举文中的两个例子，看看那些被隐藏的premises。

  1. Argument: They should never have married - they felt no strong physical attraction to each other during courtship.

给出的论点是：夫妻双方没有强烈的身体吸引力，所以他们不应该结婚，这个论点里隐藏的假定是，强烈的身体吸引力是婚姻的最佳或者唯一有意义的基础。

2. Argument: That book should be banned because it exposes children to violence.

给出的论点是：那本书向孩子们展示了暴力，应该被禁止。这里隐藏了两个假定的前提：曝光暴力是有害的，针对这种情况，禁止是最合理的反应。

最后，作者提到了段落越长的文字，隐藏的premise可能越难去辨别出来。

Whatever the length of the passenger you are evaluating , be alert for hidden premises.

前言

记得我看完这本书之后，有朋友问了我一个很常规的问题，看完有什么收获？在交流中，他提了句：follow your heart. 而我并没能清晰地向他说明为什么《beyond feelings》这本书值得去读，为什么不能一味地follow your heart，那次的沟通让我觉得虽然我看了书，但是并没有消化，准确地说，是翻完了它，不是读完了它，所以决定整理这份笔记。

第一遍看的是第6版，共22章，看得相当粗糙，想起来就翻翻，断断续续，跨时将近一年，真是头一遭，不过200多页而已……

整理的时候，重新刷了第9版，19章，相比第6版少了些，有些内容上还是变动挺大的，笔记均以第9版为准，鉴于自己蹩脚的英文语法，内容中英混杂，个人胡诌部分用「」或【】框了起来。

每章后面都附有一个Application，类似课后作业，第9版相比第6版，在application后，还新添了一个版块，A difference of opinion，我想作者本人也是很看重学以致用的能力的，惭愧的是，我基本没花什么时间在application上，故这部分内容略去。

全书共三部分，计划分四篇来呈现，一篇总览，每part单独一篇，由于第一次去整理这种非叙事性的书籍，度没控制好，每个part都是长篇，基本的思路是抽离出每章节里面的主要内容。如果你对这本书好奇，建议直接服用原版，可能货干不易消化，可配茶慢饮，如果想先大致了解下，也许我这篇总览可以作为一个参考：P。

总览

全书共三部分：

Part 1:

explains the psychological, philosophical, social context in which critical thinking takes place and describes the habits and attributes that enhance such thinking.

第一部分主要缕了下critical thinking中几个重要的基础概念：

who are you? critical thinking, truth, know, opinions, evidence, argument.

读完这部分，你会知道为什么不能一味的follow your heart。

feeling很重要，但是有时候feeling是不可靠的，feeling is notoriously unreliable，甚至我们会倾向于做一些对自己有害的事情，问问减肥的人们，多少次没有克制住去暴饮暴食？戒烟戒酒的人们，多少次认定这是最后一次开戒然后义无反顾继续smoking/drinking？明明知道这是一场很重要的考试/比赛不能搞砸，最后还是搞砸了？确认过眼神TA就是你的挚爱？你可能还需要再确认下自己是不是青光眼。

feelings need to be tested before being trusted, and thinking is the most reasonable and reliable way to test them.

你也会很惊讶于原来在critical thinker的世界里，know的定义是你不仅知道，你还知道你自己知道，并且能向人描述出你是如何知道的，而argument竟然是这样定义的：the line of reasoning that supports a judgment.

Part 2:

describes common errors in thinking and how to overcome them.

第二部分列出了在思考中容易掉进的坑，并列出针对每个坑的脱坑方法。总共24个，分别是：

errors of perspective：poverty of aspect, unwarranted assumptions, either/or outlook, mindless conformity, absolutism, relatvism, bias for or against change.

errors of procedure: biased consideration of evidence, double standard, hasty conclusion, overgeneralization and stereotyping, oversimplification, post hoc fallacy.

errors of expression: contradiction, arguing in a circle, meaningless statement, mistaken authority, false analogy, irrational appeal.

errors of reaction: automactic rejection, changing the subject, shifting the burden of proof, straw men, attacking the critic.

这些坑，或多或少都会受到一个大坑的影响，即chapter 8 单独细说的mine is better.

Part 3:

provides a step-by-step strategy for dealing with issues.

第三部分是strategy， 策略，一步步教你如何去处理一个issue。

开篇是knowing yourself，对应着本书的第一章，who are you? 这里为什么把knowing yourself放在最前面？这跟处理issue有什么关系？在处理一个issue之前，先要认清自己的不足，了解自己容易掉进part2列出的哪些坑，从而在处理issue之前就避开这些坑，随后便是being observant, 成为一个更好的观察者，一个更好的观察者才能更准确合理地去处理各种问题。chapter 16 则将关注点拉回到issues上，给出了如何去选择一个issue的方法，紧接着的三章，则是讲诉了如何去收集evidences，做出合理的interpretation，form a judgment 以及如何说服他人去接受你的这个judgment。

如果非要用一种简单的方式去展示整本书的要点，我觉得下面这张表可以提供一个非常清晰的脉络：

critical thinking的基本构成：Investigation, interpretation, judgment.

三大part基本都是围绕这个细细展开的。

一些感触

现在来回答为什么这本书值得一读。

引用书中chapter 14中的一段文字：

Noted psychologist Albert Ellis, founder of the Institute of Rational Emotive Therapy, claims, “Man can live the most self-fulfilling, creative, and emotionally satisfying life by intelligently organizing and disciplining his thinking. “

In short, although the challenge of improving your thinking is great, no other kind of self-improvement has the potential to affect every area of your life so positively.

觉得作者在夸大thinking的作用？我不觉得，至少我目前找不到反驳的论据，所以我选择暂时接受这个观点：P。

有人说，做笔记是对一本书最大的尊重。我觉得，做笔记只是一种学习方式，当你将自己从书中所学的理论或者方法，哪怕只是一点，运用于实际生活中，并因此而有所成长，才是对作者最大的尊重。因为，通过阅读他们的作品，你的生活变得更好，切切实实地有了哪怕一点微小的改变，才是一部作品能给它的受众最好的礼物。就像这本书的作者所提到的，The quality lies in the doing。

Learning is doing.

参考

Beyond feelings

写在前面

使用undef_method， 遇到的一个问题，想不明白rails是如何处理的。最后在rubyChina上发帖求助，没成想得到了很耐心的解答，贴上，以期可飨同惑者。

正文

先看一段调用undef_method的代码：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21

module MyModule
  def my_value
    "my_value in MyModule"
  end
end

class MyClass
  attr_accessor :my_value

  def initialize(value)
    @my_value = value
  end
end

class MySubClass < MyClass
  include MyModule
  undef_method :my_value
end

object = MySubClass.new(20)
puts object.my_value

执行，很明显，会报错，抛出NoMethodError

1

undefined method `my_value' for #<MySubClass:0x007fbdc4065818 @my_value=20>

没毛病，正常，ruby 2.5中对undef_method 的定义就是这样的，prevents the current class from responding to calls to the named methods。

But，你会发现rails中有一个神奇的地方，以一个简单的user model为例：

user model含有字段name， 同时include了一个module UserAddon，使用undef_method来禁止调用引入的name方法，按照undef_method的定义， 此时User的对象调用name时，应该会报错，但是进入终端后，发现没有，反而返回了正确的结果。

代码如下：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

## app/models/user.rb

# == Schema Information
#
# Table name: users
#
#  id                          :integer          not null, primary key
#  email                       :string(254)
#  name                        :string(254)
#  password_digest             :string(254)
#  created_at                  :datetime         not null
#  updated_at                  :datetime         not null

class User < ActiveRecord
  include UserAddon
  undef_method :name
end

## app/models/users_addon.rb
module UserAddon
  def name
    "hello"
  end
end

这时，rails c 进入console：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

user = User.first

# => #<User id: 1, name: "admin", email: "test@gmail.com", created_at:……>

user.methods.include?(:name)
# => false

user.name

# => "admin"

$ user.name

# => Couldn't locate a definition for user.name

user.respond_to?(:name)
# => true

使用pry的$ ，找不到user.name的定义，但是执行user.name却返回了正确的结果，此时一定调用了method_missing。但是，如何查看rails是如何实现的呢？

ruby元编程中在有关 rails的属性方法中曾经提到，当第一次访问一个属性时，这个属性是一个幽灵方法，ActiveRecord::Base # method_missing() 会把它转换成一个真实的方法，同时创建出读，写，查询方法，比如上面的name，会创建name, name=, name?，这样下次就可以直接调用。

但是，这里很显然，user.methods.include?(:name)返回了false，这里并没有创建name这个方法，那么rails是如何让user.name 返回了”admin”的呢？

这里附上来自IChou 大神的解答：

使用undef_method :name后，调用user.name, 它会落入下面的 method_missing 方法中:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23

# lib/active_model/attribute_methods.rb:425

def method_missing(method, *args, &block)
  if respond_to_without_attributes?(method, true) # =>  这里是 false
    super
  else
    match = matched_attribute_method(method.to_s)
    #<struct ActiveModel::AttributeMethods::ClassMethods::AttributeMethodMatcher::AttributeMethodMatch target="attribute", attr_name="name", method_name="name">
    match ? attribute_missing(match, *args, &block) : super
  end
end

def attribute_missing(match, *args, &block)
  __send__(match.target, match.attr_name, *args, &block)
  # match.target => "attribute"， 这里调用了自身的 attribute 方法
end

# lib/active_record/attribute_methods/read.rb:76
def _read_attribute(attr_name) # :nodoc:
  @attributes.fetch_value(attr_name.to_s) { |n| yield n if block_given? }
end
alias :attribute :_read_attribute
private :attribute

「也就是最后调用@attributes.fetch_value(attr_name.to_s) ，user.name的返回了正确的结果，而不是如预期所想的那样抛出NoMethodError。」

参考

rails attribute methods

undef method

场景：

数据迁移中遇到的一个问题，看个例子：

1
2
3
4
5
6
7
8
9

### app/models/company.rb
class Company < ActiveRecord::Base
  ……
end

### app/models/firm.rb
class Firm < Company
  ……
end

在company中，有一个字段type，按照惯例，会开启单表继承STI功能，Firm不再只是Company的一个普通的子类subclass，它与Company共享着同一个表单。

这时，需要给company添加了一个attribute，比如 creator，同时迁移数据，firm中的creator内容，需要单独调用form中的方法get_creator：

1
2
3
4
5
6
7
8
9
10
11

class AddCreatorToCompanies < ActiveRecord::Migration[5.2]
  class Company < ActiveRecord::Base; end

  def change
    add_column :companies, :creator, :string

    Comany.where(type: "Firm").find_each do |firm|
      firm.update(creator: firm.get_creator)
    end
  end
end

执行时，终端报错，显示：

1

Invalid single-table inheritance type: Firm is not a subclass of AddCreatorToCompanies::Company.

解决方法

在Company中添加 self.inheritance_column，触发单表继承STI。

1
2
3
4
5
6
7
8
9
10

class AddCreatorToComanies < ActiveRecord::Migration[5.2]
  class Company < ActiveRecord::Base
    self.inheritance_column = "_type"
  end

  def change
    add_column :companies, :creator, :string
    ……
  end
end

再次执行即可。

参考

Single table inheritance

昨晚9点不到就睡了，简直了，I want to marry my bed.

这周基本已经停了这项练习，真的是很有难度，决定不给自己平添焦虑，填上这最后一篇，暂时say bye。

• 变色的杯子，根据水温，杯子中饮品的成分，自动变换颜色，这样会让饮品变得更好喝?

• 卷寿司神器，包寿司的时候，新手容易卷不好，把海苔放进去，然后放入米饭和食材，它会自动帮你卷好寿司。

• 过敏源提醒设备，类似手环，用户输入自己的过敏源，一旦检测到用户可能在接近过敏源时，发出警告，以防用户产生过敏。【从此，远离芥末】

• 美食推荐，不同于一般的app推荐，场景如下，你在看一部电影，里面有美食场景，你可能对某款美食特别钟意，按下暂停键，截图，上传到app，它会告诉你，你所在的附近哪里有这款美食或者类似的美食，并推荐商家及用户评价，如果没有，它也会智能推荐与之味道相似的可替代食品。

• 哼歌记录。当你哼歌时，可以开启记录，它会记录下你哼过的歌，时长，音色，音调，同时记录下你爱哼的歌，每月／季度／年智能播放你的哼歌记录，相当于一个哼歌的时间记录仪。

• 夜路鼓励灯，走夜路的时候，如果害怕，可以打开这个，它除了照明，还会语音播报，告诉你前方路况，还有多久就到家了，还会播放各种笑话逗你，转移注意力。

• 漂游心情笺，每个人都可以在里面随手写一段话或者提个问题，发出去，然后他人随机阅读，每一个回复该笺的人，如果得到该笺主人的点赞，就会获得一定的奖励。

• 点子客栈，一间收集好点子的客栈，对于提出好点子的顾客，可享用美食

• 会唱歌的饮料，扫码瓶身，会播放音乐，只要你跟着音乐哼，唱完后，瓶子自动开启，享用。

• 模拟插花app，模拟插花，给到素材，自己去组合。

• 呼吸监测手链，检查你的呼吸情况，告诉你如何调整呼吸。

• 一款通过踩单车就能充电的手机，把手机放在单车的某个位置，踩单车产生的动力可以给手机充电。

• 防水的键盘，永远不用担心饮料或者茶水溅到键盘上就得去修了。

• 显示屏可以收缩的笔记本，用双手拉显示屏两边，拉大显示屏，同时保证成像精度不受损，这样就可以不用买显示器了。

• 表情包生成器应用，上传一张人脸图片，即可生成一系列基于该人脸图片的表情包。

• 屏幕清洁仪，类似女孩们用的洁面仪，不过这个是用来清洁电脑屏幕和键盘的，按一下开关，开始清扫屏幕和键盘，像是一个迷你的智能扫地机。

记录下踩过的一个坑：

类A与类B存在一对多关系，需要校验在创建b时，a存在，分别使用validates_presence_of :a 与 validates_presence_of :a_id，会发现当a尚未保存时，前者会成功创建b，后者则不行。

看个简单的例子：

group 与post 之间是一对多关系， post含字段group_id。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

### app/models/post.rb
class Post < ApplicationRecord
  belongs_to :group, optional: true
  ## 为了方便找出两者的差异，这里使用optional: true，post创建时，group_id可为nil

  validates :content, presence: true
end

#### app/models/group.rb
class Group < ApplicationRecord
  belongs_to :user
  has_many :posts

  validates :title, presence: true
end

此时，在终端执行：

1
2
3
4

post = Post.new(content: "this is a post")
post.save

# => true

接下来分别validates_presence_of :group_id和validates_presence_of :group来给post添加有效性校验，要求group必须存在。

• 在post.rb中添加validates_presence_of :group_id

  进入console，执行如下代码：

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

  post = Post.new(content: "my post") post.save #=> rollback, 返回false, Validation failed: Group can't be blank group = Group.new(title: "hello", user_id: User.first.id) post = Post.new(content: "my post") post.group = group post.save #=> rollback, 返回false post.save! #=> rollback, 显示 Validation failed: Group can't be blank

  第一次创建post，没有给到group，无法新建，正常。

  第二次创建post，使用post.group = group 给post的group赋值， 但是，你会发现虽然post.group有记录，不为nil， 但是post.group_id 依然为nil，保存时，校验group_id，为nil, 报错。

• 在post.rb中添加validates_presence_of :group

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

  post = Post.new(content: "my post") post.save #=> rollback, 返回false, Validation failed: Group can't be blank group = Group.new(title: "hello", user_id: User.first.id) post = Post.new(content: "my post") post.group = group post.save #=> 返回true, 依次在Group，Post中插入了一条新的记录

  第一次创建post，没有给到group，无法新建，正常，同validates_presence_of :group_id 的效果一样。

  第二次创建post，使用post.group = group 给post的group赋值，此时，虽然post.group_id 依然为nil，但是由于post.group不为nil，检验通过，group存在，成功创建。

  很显然，当我们创建post时，需要检验group必须存在，并不要求此时group已经创建，后者才是我们需要的校验方式。

之前看shared_context那块，对使用metadata来调用shared_context不是很理解，Google下相关文档，终于整明白了，记录下metadata的定义和使用。

是什么

官网上V3.8 的rspec-core中metadata部分，含三部分current_example, described class, User-defined metadata，但是没有明确的定义metadata[也可能是我没找到……]。

tutorials point针对metadata给到的定义：

1

data about your describe, context and it blocks.

Effective Testing with RSpec 3 在chapter8中提到Metadata解决了RSpec中的一个特殊问题：

Keep information about the context your specs are running in

类似一个信息存储盒，里面存储了如下信息：

• Example configuration(for example, marked as skipped or pending)
• Source code locations
• Status of the previous run
• How one example runs differently than others

可以把metadata看成一个带着example group相关信息的hash对象。

Metadata中默认的keys包括：

:description, :full_description, :described_class, :file_path, :example_group, :last_run_status ……..

想要查看metadata的内容，需要通过当前的example来调用metadata这个method，而example对象的引用，可以通过 it, subject, let , before, after, around 后接block来获取。

看个简单的例子感受下。

安装Rspec后，在rails项目中随意跑一个example，比如：

1
2
3
4
5
6
7
8
9

## spec/metadata_spec.rb

require 'rails_helper'

RSpec.describe :hash do
  it "hello" do |example|
    pp example.metadata
  end
end

执行后，输出如下结果：

1
2
3
4
5
6
7
8
9
10
11
12
13
14

{:block=>
  #<Proc:[email protected]/Users/xxxx/xxx/spec/metadata_spec.rb:4>,
 :description_args=>["hello"],
 :description=>"hello",
 :full_description=>"hash hello",
 :described_class=>:hash,
 :file_path=>"./spec/metadata_spec.rb",
 :line_number=>4,
 :location=>"./spec/metadata_spec.rb:4",
 :absolute_file_path=>"/Users/xxxx/xxx/spec/metadata_spec.rb",
 :rerun_file_path=>"./spec/metadata_spec.rb",
 :scoped_id=>"1:1",
 ...........
 :last_run_status=>"unknown"}

这里输出的都是metadata默认的内容，你也可以自定义metadata。

Rspec官网上举了一些例子，这里看一个简单的，给你的example加上tag fast：

1
2
3
4
5
6
7
8
9
10
11

## spec/metadata_spec.rb

require 'rails_helper'

RSpec.describe :hash do
  it "hello", fast: true do |example|
    pp example.metadata
  end
end

## 也可以写成it "hello", :fast do  |example|, Rspec默认缺省的value为true

执行后，会发现输出结果中有:fast=>true。

如果是在RSpec.describe中添加fast，则内嵌的所有example的metadata中都会带有:fast=>true。

使用场景

通过metadata可以给example或者example group添加同样的metadata，这样执行时，可以选择执行匹配要求的example了，此外也可以将module批量include到指定的example group中【后面会提到】。

那么，如何实现批量添加呢？

使用Rspec的config.define_derived_metadata。

看两个场景：

• 给某个目录下的所有example group都添加上同一个metadata
• 有条件地添加特定的metadata

指定目录下的所有example group都添加上同一个metadata

用过Rspec都知道，所有spec/controllers下的文件，都不需要添加type: :controller 这个metadata，RSpec会给所有该目录下的example group默认添加上type: :controller,这个是怎么实现的呢？

像这样：

1
2
3
4
5
6
7

### spec/spec_helper.rb

RSpec.configure do |config|
  config.define_derived_metadata(file_path: /spec\/controllers/) do |meta|
    meta[:type] = :controller
  end
end

再比如：给spec/unit下的所有example group都添加上focus这个tag。

1
2
3
4
5
6
7

### spec/spec_helper.rb

RSpec.configure do |config|
  config.define_derived_metadata(file_path: /spec\/unit/) do |meta|
    meta[:focus] = true
  end
end

有条件地添加特定的metadata

这里看一个实用性的例子，有关aggregate_failures【没用过aggregate_failures？参考部分附上了aggregate_failure的简单说明，可有助理解】。

要求除了已经添加aggregate_failures为false的example group，所有其他的example group都将aggregate_failure设置为true。

比如在clazz_a和clazz_b中，添加了aggregate_failures为false， 现在希望除了clazz_a和clazz_b外，其他都默认指定aggregate_failures为true。

1
2
3
4
5
6
7
8
9
10
11
12
13

### spec/models/clazz_a_spec.rb
RSpec.descirbe :clazz_a, aggregate_failures: false do
  context "xxx" do
    ......
  end
end

### spec/models/clazz_b_spec.rb
RSpec.descirbe :clazz_b, aggregate_failures: false do
  context "xxx" do
    ......
  end
end

实现的方式也很简单：

1
2
3
4
5
6
7

### spec/spec_helper.rb

RSpec.configure do |config|
  config.define_derived_metadata do |meta|
    meta[:aggregate_failures] = true unless meta.key?(:aggregate_failures)
  end
end

如果example group中已经添加了aggregate_failures，则pass，否则添加aggregate_failures为true。

Cool!

back to shared_context

回到之前困惑的shared_context 中。

我们知道，通过使用include_context “xxx” 可以调用定义的shared_context，同样，也可以通过给example添加metadata来调用。

比如：

1
2
3
4
5
6
7
8

#### spec/support/shared_stuff.rb

RSpec.shared_context "shared stuff" do
  let(:h) { { :hello => "world" } } ## 没啥意义的context，仅仅用来举例
end
RSpec.configure do |config|
  config.include_context "shared stuff", include_shared: true
end

使用时，只要给example group添加上include_shared: true这个metadata即可, 效果等同于调用了include_context "shared stuff"。

1
2
3
4
5
6
7
8

#### spec/shared_stuff_example.rb

require 'rails_helper'
require "support/shared_stuff.rb"

RSpec.describe "this is an example", :include_shared do
  it { expect(h[:hello]).to eq "world" }
end

执行，pass。

上面的例子，可以理解为所有添加了include_shared： true的example group 都执行了include_context "shared stuff" ,事实上，确实如此。

再看一个更常见的例子：

我们在一个module Signin中定义了一些实现不同用户登录的methods【比如sgin_in_admin, sign_in_guest……】，现在希望在spec/controller下的所有example group中都可以调用这些method的。

使用metadata可以这样实现：

1
2
3
4
5
6
7
8
9

### spec/support/sign_in.rb

module SignIn
  ..... ### methods
end

RSpec.configure do |config|
  config.include SignIn, type: :controller
end

这样，所有metadata中type为controller的example group便都会执行include SignIn了。

参考

Rspec core metadata

Rspec core shared context

Effective Testing with RSpec 3

附：

这里简单说明下aggregate_failure是干嘛用的。常规情况下，当执行一组exceptions时，在执行到第一个失败的exception后程序就跳出这个example了，不会继续执行后面的exception，直接开始执行下一个example，而通过aggregate_failure，程序会执行完该组所有的exceptions，然后列出所有failures的exceptions，再去执行下一个example。

感觉选了一个特别苦逼的练习，有时大半天都想不出一个idea，光在那里发呆，看着一片空白的typora，胡思乱想，如果有一个idea生成器该多好，然后思绪开始各种飘，也许吃点东西会更容易想出来，嗯，然后专注啃零食，也许吃点水果更健康……

祈祷下周还能整出003来。

Let ideas flying.

• 防雨鞋，下雨的时候，按一下鞋子上的按钮，会形成一层透明保护层，保证鞋不会被雨打湿

• 电量传输，手机快要没电的时候，又没有充电的地方，类似蓝牙传文件，可以使用蓝牙与另一个手机完成连接，然后传送电量

• 音乐辨识头像，根据用户的听歌记录和喜好，生成专属头像

• 肢体语言翻译app，无法言语的人，可以通过跟别人比划手势来交流，但是正常人可能并不知道如何解读，用这个肢体语言app对着别人做的手势，会翻译出对应的含义，同时录入自己想说的话，它会转换成肢体语言，给别人看【呃，既然这样，他们为什么不通过打字来交流呢？】

• 租房订阅，根据用户的设置，房间的大小，区域，价位，然后定时推送符合或者基本符合的租房信息

• 印有二维码的T恤，如果想让更多人认识你，可以让别人扫一下你的二维码，就可以获取到你的故事，了解你的喜恶。

• 代码记录，记录下你敲代码的速度和容易出错的地方，提醒你下次应该如何改进

• 坐姿提醒贴，贴在你的颈椎处，如果发现坐姿不正确，发出警告，纠正坐姿。

• 线下超市扫码买单，用户自己拿完商品，打开超市对应的app，对着商品的二维码扫一扫，商品直接添加到购物车，结算，完成买单，走人，再也不用排队了。【亚马逊有过这种结算方式，比这更高级】

• 一颗可以无限次发芽的种子，给到阳光，水，土壤就可以发芽，生长，开花，最后叶落归根，还是一颗种子，明年可以继续发芽，开花，结果，循环往复。【希望它结的是人参果】

• 红糖馒头生成器，给机器里，添加红糖和面粉，水，稍等一些时间，就可以吃到热腾腾的红糖馒头。【我一定是吃多了】

• 瘦胳膊神器，类似手套，戴上，启动，它会带动双臂开始各种摇摆，瘦胳膊同时兼减缓久坐。【减肥是人类永久的话题】

• atom快捷键智能提醒，太多的快捷键你记不住，没关系，atom记录下你的使用习惯，当你下次作出同样的操作时，弹窗，告诉你可以用哪个快捷键。

• 美景即时体验，选择你想要看的美景，比如欧洲的某个小岛，中国的某个古镇，戴上眼镜，AR即时呈现此时的那个地方，是什么样子。

• 奶茶自制机，给到茶，牛奶，糖，即可，它会自己煮好茶，然后做出一杯美味的奶茶

• 鞋子自清洁器，这个机器最好别那么大容量，是比较迷你的，使用时，放入清洁液，然后将脏脏的鞋子放进去，点击开始即可，几分钟后，鞋子干干净净。

• 不会脱落的船袜，套上去之后，不会时不时脱落。

• 先秦上古时期古文语音版，类似音乐app，但是很纯粹，没广告没评论没小视频，只有先秦各大家的作品，塞上耳机，上下班路上可以听。

背景

测试时遇到的一个问题:

新建一个对象，该对象新建后会自动保存，现在需要测试其无法保存的情景，使得obj.save返回为false。

踩坑：

obj这个对象的创建中有一个属性attr设置了必填，同时该attr的值来自另一个对象a， 参考先前的代码，解决的思路是：

通过a.save(validate: false), 让a在attr为空时保存，然后obj创建时，会取a的attr作为自己的attr，这样obj保存时必然是false，因为atrr为空，它无法过检验，但是运行后发现，不管是a.save(validate: false)还是a.save!(validate: false), 再次执行a.save，都会回滚，返回false。虽然它们跳过了validation，但是save时，并没有成功，Google了半天，没有找出好的解决方法。后来，看到同事用stub的方式完美解决了，由衷觉得真是太赞了。

同一个问题，换个人解决，高下立判，只能说，知道和想到，隔了好几个山头啊。

记录下allow_any_instance_of的用法。

allow_any_instance_of

允许一个class的任何实例stub任何一个method的，可设置返回值，也可以不设置，文档中说可以replace allow，但是allow接受的参数是对象，而allow_any_instance_of则是类。

看一些老的版本，比如2.99，会发现allow_any_instance_of的”前身”，any_instance.stub，我以为是前身，但是在该版本的Rspec-mock的message exception部分，已经出现了allow_any_instance_of，有些奇怪，不过用起来倒是都好用的，这里一并下记录下两者的用法。

• stub 单个method无参

  单个method的stub，看官网的例子：

  1 2 3 4 5 6 7 8

  ### allow_any_instance_of RSpec.describe "allow_any_instance_of" do it "returns a specified value on any instance of a class" do allow_any_instance_of(Object).to receive(:foo).and_return(:return_value) obj = Ojbect.new expect(obj.foo).to eq(:return_value) end end

  用any_instance.stub 可以这样写：

  1 2 3 4 5 6 7 8

  ### any_instance.stub RSpec.describe "any_instance.stub" do it "returns a specified value on any instance of a class" do Object.any_instance.stub(:foo).and_return(:return_value) obj = Ojbect.new expect(obj.foo).to eq(:return_value) end end

• stub单个method有参

  带参数的单个method，这样写：

  1 2 3 4 5 6 7 8 9 10 11 12

  ### allow_any_instance_of RSpec.describe "allow_any_instance_of" do context "with arguments" do it "returns the stub value when arugments match" do allow_any_instance_of(Object).to receive(:foo).with(:param_one, :param_two).and_return(:return_value_one) allow_any_instance_of(Object).to receive(:foo).with(:param_another_one, :param_another_two).and_return(:return_value_another_one) obj = Ojbect.new expect(obj.foo(:param_one, :param_two)).to eq(:return_value_one) expect(obj.foo(:param_another_one, :param_another_two)).to eq(:return_value_another_one) end end end

  换成any_instance.stub，可以这么来：

  1 2 3 4 5 6 7 8 9 10 11 12

  ### any_instance.stub RSpec.describe "any_instance.stub" do context "with arguments" do it "returns the stub value when arugments match" do Object.any_instance.stub(:foo).with(:param_one, :param_two).and_return(:return_value_one) Object.any_instance.stub(:foo).with(:param_another_one, :param_another_two).and_return(:return_value_another_one) obj = Ojbect.new expect(obj.foo(:param_one, :param_two)).to eq(:return_value_one) expect(obj.foo(:param_another_one, :param_another_two)).to eq(:return_value_another_one) end end end

• stub 多个method无参

  多个method：

  1 2 3 4 5 6 7 8 9

  ### allow_any_instance_of RSpec.describe "allow_any_instance_of" do it "stub multiple methods" do allow_any_instance_of(Object).to receive_messages(:foo => "foo", :bar => "bar") obj = Ojbect.new expect(obj.foo).to eq("foo") expect(obj.bar).to eq("bar") end end

  用any_instance.stub：

  1 2 3 4 5 6 7 8 9

  ### any_instance.stub RSpec.describe "any_instance.stub" do it "stub multiple methods" do Object.any_instance.stub(:foo => "foo", :bar => "bar") obj = Ojbect.new expect(obj.foo).to eq("foo") expect(obj.bar).to eq("bar") end end

  any_instance.stub还有一些兄弟姐妹，比如any_instance.unstub，any_instance.stub_chain， 可以戳这里了解更多。

warning

官网上提到了慎用allow_any_instance_of,原因除了认为使用该方法is often a design smell，同时也是频频出bug的地方，be careful.

参考

Stub on any instance of a class

any instance