思考与写作paulgraham.com编辑精选
如何提出异议
How to Disagree
2008 — 2011，它一次次回到首页
黑点是上首页（≥ 50 分）的那次投稿，空心点是没冲上去的投稿，点越大分数越高。
导读
Paul Graham 的这篇文章提出“反驳层级”，从 DH0 的辱骂、人身攻击、挑语气措辞，到 DH4 的相反观点、DH5 的反驳，再到最高级的 DH6 直接反驳核心论点。它用简单框架帮助读者识别网络争论的质量，并提醒人们把精力放在论证本身。
为什么经久不衰：它提供了一个清晰、可操作的分级工具，让人能立刻判断一场争论是在攻击人还是在讨论问题，因此多年后仍值得重读。
HN 评论精选
以阿奎那为例说明最高层级反驳需要谦逊追求真理，并先强化对方论证。
好文章。Paul 对这个问题做了有用的阐明。中世纪神学家托马斯·阿奎那常常在 DH6 层面运作，效果惊人：他会清晰、深入、准确、精确地表述自己打算反驳的论证，甚至常把对手的论点说得比对手本人更好。我相信他的秘诀是谦逊地关心真理，无论它通向何处；因为没有预设议程，他能把一个命题推进很深，在每一步都为它做最好的论证，然后再收回并有力地陈述结论。谦逊、抽离、无议程地倾听并追随真理，以及在找到艰难获得的真理后有力捍卫它，似乎都是很好的起点。
gregfjohnson查看英文原文
Great article. Paul provides a useful illumination of the issue. The great medieval theologian Thomas Aquinas operated at level DH6, often with devastating effect. He would formulate and state the argument he planned to refute with clarity, depth, accuracy, and precision. He often would make his opponent's case better than the opponent had done! I believe his secret was that he was humbly interested in the truth wherever it led. Given his lack of a pre-conceived agenda, he could take a proposition and explore its implications 20 plies deep, giving it the benefit of the doubt and making the best case for it at each step. Having fully plumbed an idea and 'grokked' it, then he could unwind the stack as it were, and state his conclusions in a powerfully convincing manner. Humility, detachment, an agenda-free willingness to listen and follow the truth where it leads, and strength in defending hard-won truth when found all seem to be good places to start.
提出 DH-1“引战”：用片面结论诱导低层级反驳。
介意我扩展你的反驳层级吗？DH-1：引战。这是指你预先通过以会引发快速反对（DH0 到 DH6）的方式陈述观点，来先发制人地应对你将不同意的内容。通常做法是只注意全部数据的一个子集，形成一个只解释该子集的假设，并把它当作结论提出，常常甚至没有意识到。例如：关系数据库模型已死；企业软件没有未来；某某当红语言显然最好；没有别人做某件事，所以它一定是死路。
edw519查看英文原文
Mind if I extend your Disagreement Hierarchy? DH-1. Flamebaiting. This is when you preempt that with which you would disagree by stating your case in such a way as to elicit quick disagreement (in the form of DH0 to DH6). Usually done by noticing a subset of all data, forming a hypothesis to explain only that subset, and presenting it as a conclusion. Often done without even realizing it. The relational database model is dead. There is no future in enterprise software. <Language du jour> is clearly the best. No one else is doing <xyz>, so it must be a deadend.
用亚里士多德三要素说明攻击信誉或共情有时也正当，取决于论证是否依赖它们。
很好的文章，谢谢！有个想法：根据亚里士多德《修辞学》，向听众表达观点的人有三样东西：Ethos 信誉，即他是谁、为何有权威；Pathos 共情，即与听众的情感共鸣；Logos 逻辑，即论证实质。因此对手可以试图削弱三者之一。削弱信誉可以是粗俗的人身攻击，也可以更精巧；如果对方论证依赖其信誉，那可能正当。例如诺贝尔奖得主 William Shockley 凭物理学背景主张种族优生学，说“教授，您的专长是物理学，不是遗传学”就是正当的。削弱共情也可能有帮助，比如提醒听众某软件营销人员是想多卖许可证。当然，削弱逻辑是非常有效的反驳方式；但我的观点是，削弱说话者的共情和信誉也可能正当，尤其当论证关键依赖这些方面时。
JoeBackward查看英文原文
Very nice post, thank you! Here's something to think about. According to Aristotle's Rhetoric, a person making a point to an audience has three things to offer: -- Ethos -- who they are and why they speak with authority. -- Pathos -- empathy with the audience. -- Logos -- the substance of the argument. So, an opponent can try to undermine any of these three. Undermining a person's ethos can be nasty "yo' momma" style language, or it can be more sophisticated. If the person's argument relies on their ethos, however, it may be legitimate. For example, when Nobel laureate William Schockley argues for racial eugenics on the strength of his physics background, it's legitimate to say "Professor, your expertise is in physics, not genetics." This is helpful with anyone who says, effectively, "I'm a bigshot so what I say must be true." Undermining pathos can also be helpful. For example, "thus-and-such software marketer doesn't want to help you and me, she wants to sell more software licenses for her company. Do not blindly accept her claims that her product is better." Of course, undermining the logos -- the substance of the argument -- is a very effective way to disagree. But, my point is, undermining a speaker's pathos and ethos are also legitimate, especially when their argument critically depends on those aspects of their rhetoric.
提出 DH7：先补强到对方最佳版本并提炼核心，再驳倒。
好文章。不确定是否与 DH6 有显著不同，但卡尔·波普尔的辩论风格或可称为 DH7。它类似 DH6，但你先修补对手的论证，做出最好版本；然后找到该论证的核心点，即使作者没有明确说出。那是他真正想说的，若状态好可能也会说的。如果有更强或更一般的版本，你也可以选那个。然后你仔细摧毁它。
tungstenfurnace查看英文原文
Nice article. Not sure if it's significantly different from DH6, but Karl Popper's style of debate might be termed DH7. It's similar to DH6, but first you patch up the opponent's arguments to make the best possible case. Then you find the central point in that case, even if the author doesn't explicitly state it. It's what he really meant to say and might indeed have said if he was on good form. (If there's a stronger or more general version of that point you might select that instead.) And _then_ you carefully demolish it.
历次讨论
同类还值得读
保持你的身份认同足够小
Keep Your Identity Small (2009)
Paul Graham 2009 年的一篇短文：人一旦把某个话题当成自己身份的一部分（宗教、政治、编程语言皆然），讨论就会退化成站队和争吵，因为此时争论的不再是事实，而是"我是谁…
思考与写作上首页 5 次 · 2009–2021paulgraham.com编辑精选苏格拉底式教学法：通过提问而非讲授来教学
The Socratic Method: Teaching by Asking Instead of by Telling
记录了作者用纯提问的方式，引导一群小学生从零推导出二进制计数法的课堂实验，展示了苏格拉底式教学法的惊人效果。
思考与写作上首页 5 次 · 2011–2022garlikov.com你不能说的话
What You Can't Say (2004)
Paul Graham 2004 年的经典文章，探讨那些因道德时尚而不能公开说出的观点。
思考与写作上首页 5 次 · 2014–2022paulgraham.com编辑精选龙暴君的寓言（2005）
The Fable of the Dragon-Tyrant (2005)
哲学家 Nick Bostrom 于 2005 年写的一则寓言：一条每天吞噬数千人的巨龙（隐喻死亡与衰老）长期被人类当作不可抗拒的自然规律接受，直到有人提出并最终用导弹将其杀死，…
思考与写作上首页 5 次 · 2009–2021nickbostrom.com编辑精选现实有着惊人的细节量
Reality has a surprising amount of detail (2017)
2017 年的博文指出，现实远比我们以为的充满细节：我们习惯用简化的心智模型看世界，直到真正动手做事（比如作者举的造楼梯例子）才发现处处是未曾预料的细节。
思考与写作上首页 4 次 · 2018–2025johnsalvatier.org编辑精选