Concise Summary简洁概述
Yudkowsky draws a sharp algorithmic line between two mental processes that superficially resemble each other. Rationality is a forward flow: gather evidence, weigh it, arrive at a conclusion. Rationalization is a backward flow: start with the conclusion already written, then search for evidence to justify it. These are not the same process run in opposite directions — they are computationally distinct operations with opposite epistemic effects. Rationality can change your beliefs and bring them closer to truth; rationalization locks beliefs in place and dresses them in borrowed arguments. The essay also indicts Traditional Rationality for tolerating the backward flow, and closes with the claim that curiosity — not knowing your destination — is the first virtue of genuine inquiry.
Yudkowsky 在两种表面相似的心理过程之间划出一条清晰的算法界线。理性是正向流动：收集证据，权衡，得出结论。合理化是反向流动：先写下既定结论，再寻找证据为其背书。这两者并非同一过程的两个方向——它们在计算上是不同的操作，产生截然相反的认识论效果。理性能改变信念，使其更接近真相；合理化则把信念锁死原地，再给它披上借来的论证。本文还批判了传统理性主义对这种反向流动的容忍，并以「好奇心——不知道自己的目的地——是真正探究的第一美德」作结。
Infographic信息图
The curious inquirer vs. the clever arguer
好奇的探究者 vs. 聪明的论证者
The curious inquirer lists all evidence and flows forward to a probability. The clever arguer writes the conclusion first, then selects evidence to fill the lines above it.
好奇的探究者列出所有证据，正向推导出概率。聪明的论证者先写好结论，再从中筛选证据填满结论上面的行。
Rationalization: a cognitive black hole
合理化：认知黑洞
Yudkowsky's preferred name for rationalization is "giant sucking cognitive black hole" — a process that fixes beliefs rather than updating them.
Yudkowsky 更愿意把合理化称为「巨大的认知吸力黑洞」——一个固定信念而非更新信念的过程。
Arguing cannot make truth
论证无法制造真相
You can make more people believe a proposition by arguing it, but you cannot make the proposition more true. Only changing beliefs, not reinforcing them, can improve accuracy.
你可以通过论证让更多人相信某个命题，但无法让这个命题变得更真实。只有改变信念，而非强化信念，才能提高准确性。
Curiosity as the first virtue
好奇心是第一美德
In genuine rationality you do not know your destination; that's what makes you curious. If you already know where you're going, in Bayesian terms you're already there.
在真正的理性中，你不知道自己的目的地——这正是你感到好奇的原因。用贝叶斯的话说：如果你已经知道要去哪里，你就已经到了那里。
Traditional Rationality's blind spot
传统理性主义的盲点
Traditional Rationality praises the scientist who forms a hypothesis and then designs an experiment to prove it. Yudkowsky argues this tolerates the backward flow and would be better replaced by experiments driven by curiosity alone.
传统理性主义称赞那种先提出假说、再设计实验来证明它的科学家。Yudkowsky 认为这容忍了反向流动，更好的做法是纯粹出于好奇心去驱动实验。
Detailed Summary详细概述
The Setup: Two Boxes, Two Procedures
Yudkowsky opens by recalling his earlier essay "The Bottom Line," where a diamond is hidden in one of two boxes and various signs and portents serve as evidence. He contrasts two characters: the curious inquirer, who writes down all visible signs and processes them forward to arrive at a probability like "85% chance box B contains the diamond"; and the clever arguer, who begins by writing the conclusion—"box B contains the diamond"—and then selects favorable signs to list above it.
The first procedure is rationality. The second is what is commonly called "rationalization."
The Word Problem
Yudkowsky pauses on the word itself. "Rationalization" is, in his view, a wrong word. You cannot "rationalize" what is not already rational — the term suggests a process that produces rationality, when in fact it destroys it. He offers a pointed analogy: it would be like calling lying "truthization." His preferred alternative is "giant sucking cognitive black hole" — a name that makes the algorithmic difference obvious.
The Computational Distinction
At the core of the essay is a precise distinction between two computational procedures:
- Forward flow (rationality): Start from evidence → crunch probabilities → output a conclusion. The destination is unknown at the start.
- Backward flow (rationalization): Start from a fixed conclusion → crunch backward → select evidence that supports it. The destination is known; only the supporting arguments are variable.
These are not the same process run in reverse. They have opposite epistemic effects. Rationality is designed to update beliefs toward accuracy. Rationalization is designed to hold beliefs fixed — it is, as Yudkowsky puts it, "anti-rationality," both in its pragmatic results and in its reversed algorithm.
Not every change is an improvement, but every improvement is necessarily a change.
You cannot make a proposition more true by arguing for it. You can increase the number of people who believe it, but truth is not a property you can add to a fixed belief. Only by allowing beliefs to change can you improve their accuracy.
Curiosity and the Bayesian Point
Yudkowsky emphasizes that the curious inquirer, while running the forward-flow algorithm, does not yet know their destination — and that is precisely why they are curious. In the Way of Bayes, the prior probability equals the expected posterior probability: if you already know where you are going, you are already there. A process that begins by writing the conclusion is not in any meaningful sense still "inquiring."
Traditional Rationality's Failure
The essay turns critical. Traditional Rationality does not adequately distinguish forward from backward flow. It praises the scientist who arrives at a "pet hypothesis" and then designs an experiment to prove it — calling this pride "the engine that drives Science forward." Yudkowsky concedes the engine analogy has some truth (it is easier to find opposed advocates than a single unbiased inquirer), but insists this does not make backward-flow reasoning acceptable. It would be better if the scientist, upon forming a hypothesis, sought to test it out of genuine curiosity — designing experiments whose results would push their beliefs in an unknown direction.
The Closing Exhortation
The essay ends with a spare, memorable line: Feel the flow of the Force, and make sure it isn't flowing backwards. Curiosity is named the first virtue, without which questioning is purposeless and skill is directionless. The entire argument reduces to one diagnostic: when you are thinking about a topic, is the unknown thing the conclusion, or the supporting arguments for a conclusion you've already reached?
开场：两个盒子，两种程序
Yudkowsky 开篇回顾了他此前的文章《底线》：一颗钻石藏在两个盒子之一中，各种迹象作为证据。他对比了两个人物：好奇的探究者，写下所有可见迹象，正向处理，得出「85% 概率盒子 B 含有钻石」这样的结论；以及聪明的论证者，先写下结论——「盒子 B 含有钻石」——再从中筛选有利迹象列在结论上方。
第一种程序是理性。第二种通常被称为「合理化」。
词语问题
Yudkowsky 在这个词本身上停顿了一下。在他看来，「合理化」（rationalization）是一个错误的词。你无法将本不理性的东西「合理化」——这个词暗示一个产生理性的过程，而实际上它摧毁理性。他给出一个尖锐的类比：这就像把「撒谎」称为「真相化」。他更倾向的替代词是「巨大的认知吸力黑洞」——一个能让算法差异一目了然的名字。
计算上的区分
本文核心是两种计算程序之间的精确区分：
- 正向流动（理性）： 从证据出发 → 计算概率 → 输出结论。起点时目的地未知。
- 反向流动（合理化）： 从固定结论出发 → 反向计算 → 筛选支持它的证据。目的地已知，变量只是支持性论证。
这两者并非同一过程的正反向运行——它们产生截然相反的认识论效果。理性旨在将信念更新为更准确的状态；合理化旨在将信念固定原地——Yudkowsky 称之为「反理性」，无论从实际结果还是反向算法上看，皆然。
并非所有改变都是进步，但所有进步必然是改变。
你无法通过论证让一个命题变得更真实。你可以增加相信它的人数，但真相不是可以附加给一个固定信念的属性。只有允许信念改变，才能提高其准确性。
好奇心与贝叶斯要点
Yudkowsky 强调，好奇的探究者在运行正向流动算法时，尚不知道自己的目的地——这正是他感到好奇的原因。在贝叶斯之道中，先验概率等于期望后验概率：如果你已经知道要去哪里，你就已经到了那里。 一个从写下结论开始的过程，在任何有意义的层面上都称不上「探究」。
传统理性主义的失败
文章转向批判。传统理性主义未能充分区分正向与反向流动。它称赞那种提出「心爱假说」然后设计实验来证明它的科学家，称这种自豪感是「推动科学前进的引擎」。Yudkowsky 承认这个引擎类比有一定道理（找到两个各执一方的辩护者，比找到一个无偏见的探究者更容易），但坚持认为这并不使反向流推理变得可接受。更好的做法是：科学家在形成假说后，出于真正的好奇心去检验它——设计那些结果会将信念推向未知方向的实验。
结语告诫
文章以一句简洁而令人难忘的话作结：感受原力的流动，确保它没有倒流。 好奇心被称为第一美德，没有它，追问便无目的，技能便无方向。整个论证归结为一个诊断问题：当你思考某个话题时，那个未知的东西是结论，还是你已得出的结论所需要的支持性论证？
FAQ常见问答
What is the precise difference between rationality and rationalization?理性与合理化之间的精确区别是什么？
It's an algorithmic difference, not just a matter of degree. Rationality flows forward: evidence → probability calculation → conclusion (unknown in advance). Rationalization flows backward: conclusion (fixed in advance) → selection of supporting evidence. The destination in rationality is the variable; in rationalization the destination is constant and only the supporting arguments vary.
这是一种算法上的差异，而非程度上的差别。理性正向流动：证据 → 概率计算 → 结论（事先未知）。合理化反向流动：结论（事先固定）→ 筛选支持性证据。在理性中，目的地是变量；在合理化中，目的地是常量，只有支持性论证是变量。
Can't you reach true conclusions through rationalization, just by luck?难道不能通过合理化，凭运气得出真实结论吗？
Yes, occasionally — but that's not the point. The essay's concern is procedural reliability, not any single outcome. A backward-flowing process that happens to land on a true conclusion got there by fixing the conclusion first, not by evidence. It cannot systematically improve accuracy, and it offers no way to detect when it's wrong.
偶尔可以——但这不是重点。文章关注的是程序上的可靠性，而非某一次结果。一个恰好得出真实结论的反向流动过程，之所以到达那里，是因为它先固定了结论，而非依靠证据。它无法系统性地提高准确性，也无法检测自身何时出错。
Why does Yudkowsky call rationalization "anti-rationality" rather than just "bad reasoning"?为什么 Yudkowsky 把合理化称为「反理性」而不仅仅是「糟糕的推理」？
Because it's not merely reasoning done poorly — it is reasoning's functional opposite. Rationality operates to update beliefs toward accuracy; rationalization operates to fix beliefs in place. "Bad reasoning" implies a failed attempt at the right goal; "anti-rationality" names a process with the opposite goal: locking down a predetermined conclusion rather than discovering one.
因为它不仅仅是推理做得不好——它是推理在功能上的对立面。理性的运作目的是将信念更新为更准确的状态；合理化的运作目的是将信念固定原地。「糟糕的推理」意味着对正确目标的失败尝试；「反理性」命名的是一个目标相反的过程：锁定一个预设结论，而非去发现结论。
What's wrong with the scientist who forms a hypothesis and then tests it?先提出假说再去检验它的科学家，有什么问题？
Yudkowsky's concern is subtle: the problem isn't hypothesis formation, it's motivation. A scientist driven by the desire to prove a hypothesis will design experiments biased toward confirmation. A scientist driven by genuine curiosity about whether the hypothesis is true will design experiments that could go either way — pushing their beliefs in an unknown direction. The first is backward flow; the second is forward flow.
Yudkowsky 的担忧很微妙：问题不在于形成假说，而在于动机。一个想要证明某假说的科学家会设计出偏向确认的实验。一个真正好奇于假说是否成立的科学家，会设计出可能指向任一方向的实验——将信念推向未知的方向。前者是反向流动，后者是正向流动。
How does the Bayesian point about priors connect to curiosity?关于先验的贝叶斯要点与好奇心有何关联？
In Bayesian terms, your prior over a question represents your genuine uncertainty before seeing evidence. If you "already know" the answer — if the conclusion is fixed before inquiry begins — then your prior is effectively 1 or 0, and no evidence can update it. Curiosity, by contrast, is the psychological correlate of having a real prior with genuine uncertainty. You are curious precisely because you don't know the answer yet.
用贝叶斯的语言说，你对某个问题的先验代表着你在看到证据之前的真实不确定性。如果你「已经知道」答案——如果结论在探究开始之前就已固定——那么你的先验实际上是 1 或 0，任何证据都无法更新它。好奇心恰恰相反，它是拥有真实先验、存在真实不确定性的心理对应物。你之所以感到好奇，正是因为你还不知道答案。
Does this essay apply to everyday thinking, or only to formal reasoning?这篇文章适用于日常思维，还是只适用于正式推理？
Yudkowsky explicitly intends it for everyday thinking. The diagnostic question is simple: when you're reasoning about a topic, is the unknown thing the conclusion, or the arguments you need to support a conclusion you've already reached? If the latter, the flow is backwards — whether you're deciding which box holds a diamond, evaluating a political claim, or reconsidering a personal belief.
Yudkowsky 明确地将其用于日常思维。诊断问题很简单：当你思考某个话题时，那个未知的东西是结论，还是你需要用来支持一个已达成结论的论证？如果是后者，流动方向就是反向的——无论你是在判断哪个盒子里有钻石、评估一个政治主张，还是重新审视一种个人信念。
In-depth Analysis · Pros & Cons深入解读 · 优缺点
"Rationalization" is one of Yudkowsky's most precise short essays, making a single algorithmic point with unusual clarity. Its contribution is to reframe a familiar cognitive bias not as a moral failing but as a computational distinction — two processes that look similar from the outside but differ in what is held fixed and what is allowed to vary.
《合理化》是 Yudkowsky 最精准的短文之一，以异常清晰的方式阐明了一个算法要点。它的贡献在于将一种熟悉的认知偏差重新定义为一种计算上的区分——两个从外部看起来相似的过程，但在「什么被固定」和「什么被允许变化」上截然不同。
- Algorithmic precision算法上的精准By framing rationalization as a reversed computational procedure rather than as motivated reasoning or wishful thinking, Yudkowsky gives readers a crisp, checkable diagnostic: which element is the unknown — the conclusion or the supporting arguments?通过将合理化定义为一种反向计算程序，而非动机性推理或一厢情愿，Yudkowsky 给读者提供了一个清晰、可检验的诊断标准：哪个要素是未知的——结论还是支持性论证？
- The word critique is genuinely useful对词语的批判真正有用Pointing out that "rationalization" misleadingly suggests a rationality-producing process — and that "anti-rationality" or "cognitive black hole" would be more accurate — is not mere wordplay; it highlights how language can obscure the severity of a cognitive failure.指出「合理化」这个词误导性地暗示了一个产生理性的过程——而「反理性」或「认知黑洞」会更准确——这不是玩文字游戏；它揭示了语言如何能掩盖认知失败的严重性。
- Curiosity as positive program好奇心作为积极纲领Rather than merely diagnosing a failure mode, the essay ends with a positive prescription: cultivate curiosity — the state of not knowing your destination — as the practical safeguard against backward flow. This gives the diagnosis a constructive counterpart.文章不仅仅是在诊断一种失败模式，而是以积极的处方作结：培养好奇心——那种不知道自己目的地的状态——作为对抗反向流动的实践保障。这给诊断提供了一个建设性的对应物。
- The Traditional Rationality critique is pointed对传统理性主义的批判切中要害Conceding that backward-flow reasoning is what drives actual scientific progress, then arguing it isn't optimal, is a more intellectually honest move than simply condemning it — it shows awareness of the practical constraints scientists face.承认反向流推理确实是驱动实际科学进步的东西，然后论证它并不是最优的，比简单地谴责它更诚实——这表明作者意识到了科学家面临的实际约束。
- The forward/backward distinction can blur in practice正向/反向的区分在实践中可能模糊Real reasoning rarely starts from either pure evidence or a pure conclusion. Scientists often work iteratively — forming a tentative hypothesis, gathering evidence, revising the hypothesis, gathering more evidence. The essay's clean binary may not carve cognition at its joints; partial forward-flow mixed with partial backward-flow is the common case.真实的推理很少从纯粹的证据或纯粹的结论出发。科学家通常是迭代工作的——形成初步假说，收集证据，修正假说，再收集更多证据。文章干净的二元区分可能并没有切中认知的关节处；正向流与反向流的部分混合才是常见情况。
- Curiosity is easier to prescribe than to produce好奇心开出来容易，培养出来难Saying "be curious" and "don't know your destination" is correct advice but epistemically shallow. Strong motivated reasoning is precisely what makes genuine curiosity hard to access — you often can't easily tell whether your sense of openness is genuine or performed. The essay gives no guidance on how to produce the curious state when rationalization is already running.说「保持好奇」和「不要知道自己的目的地」是正确的建议，但在认识论上流于浅表。强烈的动机性推理恰恰是让真正的好奇心难以进入的原因——你往往很难判断自己的开放感是真实的还是表演的。文章没有提供任何指导，说明当合理化已经在运行时，如何重新进入好奇状态。
- "You cannot add truth to a fixed proposition" conflates truth and credence「无法为固定命题添加真值」混淆了真值与置信度Yudkowsky's claim that arguing a proposition cannot make it "more true" is technically correct — truth is binary — but the relevant epistemic question is about credence, not truth. The essay's phrasing risks obscuring the real point, which is that backward-flow reasoning fails to track probability accurately, not that it somehow alters an external fact.Yudkowsky 关于论证无法让命题「更真实」的说法在技术上是正确的——真值是二元的——但相关的认识论问题是关于置信度，而非真值。文章的措辞有混淆真正要点的风险：真正的要点是反向流推理无法准确追踪概率，而非它以某种方式改变了外部事实。
- The scope of the indictment is wider than the argument supports指控的范围比论证所支持的更宽The essay condemns backward-flow reasoning quite broadly, but some reasoning that starts from a conclusion is legitimate — e.g., a defense attorney whose job is to advocate for a predetermined client, or an engineer stress-testing a design by actively seeking failure modes. The normative target (when backward flow is cognitive rather than institutional) needs sharper specification.文章相当广泛地谴责了反向流推理，但某些从结论出发的推理是正当的——例如，辩护律师的职责本来就是为预定委托人辩护，或者工程师通过主动寻找失败模式来进行压力测试。规范目标（反向流动在认知层面而非制度层面何时有问题）需要更清晰的界定。
A compact and well-aimed essay that does exactly what it sets out to do: replace a vague moral intuition ("rationalization is bad") with a crisp algorithmic diagnosis ("conclusion-first processing cannot update beliefs toward truth"). Its limits are those of any short polemical piece — the neat binary framing glosses over the messy iterative reality of most human reasoning, and the prescription ("be curious") points in the right direction without providing the mechanics. Read as a conceptual tool for identifying backward flow, it is excellent; read as a complete guide to avoiding it, it leaves most of the work undone.
这是一篇简洁而目标精准的文章，完成了它设定的任务：用一个清晰的算法诊断（「结论优先的处理无法将信念更新为真相」）取代模糊的道德直觉（「合理化是坏事」）。它的局限性与任何短小论战文章的局限性相同——整洁的二元框架掩盖了大多数人类推理中混乱迭代的现实，而处方（「保持好奇」）指向了正确方向，却没有提供具体机制。作为一种用于识别反向流动的概念工具，它非常出色；作为避免反向流动的完整指南，它留下了大部分工作未完成。
Original Text原文
In “The Bottom Line,” I presented the dilemma of two boxes, only one of which contains a diamond, with various signs and portents as evidence. I dichotomized the curious inquirer and the clever arguer. The curious inquirer writes down all the signs and portents, and processes them, and finally writes down, “Therefore, I estimate an 85% probability that box B contains the diamond.” The clever arguer works for the highest bidder, and begins by writing, “Therefore, box B contains the diamond,” and then selects favorable signs and portents to list on the lines above.
The first procedure is rationality. The second procedure is generally known as “rationalization.”
“Rationalization.” What a curious term. I would call it a wrong word. You cannot “rationalize” what is not already rational. It is as if “lying” were called “truthization.”
On a purely computational level, there is a rather large difference between:
- Starting from evidence, and then crunching probability flows, in order to output a probable conclusion. (Writing down all the signs and portents, and then flowing forward to a probability on the bottom line which depends on those signs and portents.)
- Starting from a conclusion, and then crunching probability flows, in order to output evidence apparently favoring that conclusion. (Writing down the bottom line, and then flowing backward to select signs and portents for presentation on the lines above.)
What fool devised such confusingly similar words, “rationality” and “rationalization,” to describe such extraordinarily different mental processes? I would prefer terms that made the algorithmic difference obvious, like “rationality” versus “giant sucking cognitive black hole.”
Not every change is an improvement, but every improvement is necessarily a change. You cannot obtain more truth for a fixed proposition by arguing it; you can make more people believe it, but you cannot make it more true. To improve our beliefs, we must necessarily change our beliefs. Rationality is the operation that we use to obtain more accuracy for our beliefs by changing them. Rationalization operates to fix beliefs in place; it would be better named “anti-rationality,” both for its pragmatic results and for its reversed algorithm.
“Rationality” is the forward flow that gathers evidence, weighs it, and outputs a conclusion. The curious inquirer used a forward-flow algorithm: first gathering the evidence, writing down a list of all visible signs and portents, which they then processed forward to obtain a previously unknown probability for the box containing the diamond. During the entire time that the rationality-process was running forward, the curious inquirer did not yet know their destination, which was why they were curious. In the Way of Bayes, the prior probability equals the expected posterior probability: If you know your destination, you are already there.
“Rationalization” is a backward flow from conclusion to selected evidence. First you write down the bottom line, which is known and fixed; the purpose of your processing is to find out which arguments you should write down on the lines above. This, not the bottom line, is the variable unknown to the running process.
I fear that Traditional Rationality does not properly sensitize its users to the difference between forward flow and backward flow. In Traditional Rationality, there is nothing wrong with the scientist who arrives at a pet hypothesis and then sets out to find an experiment that proves it. A Traditional Rationalist would look at this approvingly, and say, “This pride is the engine that drives Science forward.” Well, it is the engine that drives Science forward. It is easier to find a prosecutor and defender biased in opposite directions, than to find a single unbiased human.
But just because everyone does something, doesn’t make it okay. It would be better yet if the scientist, arriving at a pet hypothesis, set out to test that hypothesis for the sake of curiosity—creating experiments that would drive their own beliefs in an unknown direction.
If you genuinely don’t know where you are going, you will probably feel quite curious about it. Curiosity is the first virtue, without which your questioning will be purposeless and your skills without direction.
Feel the flow of the Force, and make sure it isn’t flowing backwards.
在《底线》中，我提出了一个两个盒子的难题，其中只有一个装有钻石，各种迹象作为证据。我将好奇的探究者与聪明的论证者做了二分对比。好奇的探究者写下所有迹象与预兆，对它们进行处理，最终写下：「因此，我估计盒子 B 含有钻石的概率为 85%。」聪明的论证者为出价最高者效力，他先写下「因此，盒子 B 含有钻石」，然后从中挑选有利的迹象与预兆，列在上面的各行。
第一种程序是理性。第二种程序通常被称为「合理化」（rationalization）。
「合理化」。多么奇怪的词。我会称它为一个错误的词。你无法将本不理性的东西「合理化」。这就好比把「撒谎」称为「真相化」（truthization）一样。
从纯粹计算的层面来看，以下两者之间存在相当大的差异：
- 从证据出发，计算概率流，从而输出一个可能的结论。（写下所有迹象与预兆，然后正向推导，得出依赖于那些迹象与预兆的底线概率。）
- 从结论出发，计算概率流，从而输出表面上支持该结论的证据。（写下底线，然后反向推导，筛选迹象与预兆用于在上方各行展示。）
什么蠢货想出了「理性」（rationality）和「合理化」（rationalization）这两个令人迷惑的相似词语，来描述如此截然不同的心理过程？我更希望用能让算法差异一目了然的词，比如「理性」对「巨大的认知吸力黑洞」。
并非所有改变都是进步，但所有进步必然是改变。你无法通过论证一个固定命题来为其增加更多真实性；你可以让更多人相信它，但你无法让它变得更真实。要改善我们的信念，我们就必须改变我们的信念。理性是我们通过改变信念来获得更高准确性的操作。合理化则将信念固定于原地；它更应该被命名为「反理性」，无论是从其实际结果还是从其反向算法来看，皆然。
「理性」是正向流动，它收集证据，权衡证据，输出结论。好奇的探究者使用了一种正向流动算法：首先收集证据，写下所有可见迹象与预兆的列表，然后正向处理它们，得出一个之前未知的、关于盒子含有钻石的概率。在整个理性过程正向运行的过程中，好奇的探究者尚不知道自己的目的地，这正是他们感到好奇的原因。在贝叶斯之道中，先验概率等于期望后验概率：如果你已经知道自己的目的地，你就已经到了那里。
「合理化」是从结论到精选证据的反向流动。你首先写下底线，它是已知且固定的；你处理过程的目的是找出你应该在上方各行写下哪些论证。这才是运行中的过程所不知道的变量，而非底线。
我担心传统理性主义并没有使其用户对正向流动与反向流动之间的差异足够敏感。在传统理性主义中，一个科学家形成了一个心爱的假说然后着手寻找实验来证明它，这没有任何问题。传统理性主义者会赞许地看待这一切，说：「这种自豪感是推动科学前进的引擎。」好吧，它确实是推动科学前进的引擎。找到两个偏向相反方向的检察官和辩护者，比找到一个毫无偏见的人要容易得多。
但仅仅因为每个人都做某事，并不意味着它就是对的。如果那位科学家，在形成一个心爱假说之后，出于好奇心去检验那个假说——创造能将自己的信念推向未知方向的实验——那将会更好。
如果你真的不知道自己要去哪里，你大概会对此感到相当好奇。好奇心是第一美德，没有它，你的追问便毫无目的，你的技能便失去方向。
感受原力的流动，确保它没有倒流。