Concise Summary简洁概述
In our ancestral hunter-gatherer bands, useful knowledge was universally shared: you were never more than one inferential step from your listener. Modern disciplines can demand a hundred inferential steps from shared premises to a conclusion — but our social instincts never updated. So explainers take only one step back when they need two or three, and listeners brand an unsupported claim as crazy rather than recognizing that the necessary background simply has not been established. The fix is deliberate: lay out a full inferential pathway starting from what your audience actually knows, and never assume that a term like "simplest explanation" carries the same epistemic weight for them that it carries for you.
在我们祖先的狩猎采集族群里，有用的知识是普遍共享的：你与听众之间从不超过一步的推理距离。现代学科的结论可能与共同前提相距一百个推理步骤——但我们的社会本能从未更新。于是，解释者在需要退两三步时只退了一步，而听众则把没有明显依据的说法直接定性为「疯了」，而非意识到必要的背景知识根本还未建立。解决之道是刻意的：从听众实际已知的内容出发，铺陈完整的推理路径，并且绝不假设「最简解释」这样的词汇对他们具有与对你同等的认识论分量。
Infographic信息图
The ancestral baseline
祖先的基准
In a band of 200 with no writing, all background knowledge was public knowledge. You only ever needed to share one new piece of information.
在一个没有文字、人数不超过200的族群里，所有背景知识都是公共知识。你从来只需要分享一条新信息。
The unchanged instinct
未被更新的本能
Our brains still expect short inferential distances. An unsupported claim triggers "liar or idiot," not "this person has background I lack."
我们的大脑仍然预期推理距离很短。一个缺乏明显依据的说法会触发「说谎或蠢货」，而非「这个人有我所不具备的背景知识」。
The biologist–physicist gap
生物学家与物理学家之间的鸿沟
"It's the simplest explanation" lands differently on someone steeped in Occam's Razor versus someone who hasn't been. The same words, radically different inferential weight.
「这是最简解释」，对一个深受奥卡姆剃刀熏陶的人与对一个未曾接触过它的人，分量截然不同。同样的词语，截然不同的推理分量。
Laying the pathway
铺陈推理路径
A clear argument must start from what the audience already knows and walk each inferential step. Skip a step and you're talking to yourself.
一个清晰的论证必须从听众已知的东西出发，逐步走过每一个推理步骤。跳过一步，你就只是在自言自语。
The condescension trap
居高临下的陷阱
Don't let your audience sense you think you're a dozen steps ahead of them. That reads as arrogance, not expertise.
别让听众感觉到你认为自己比他们领先十几步。那会被解读为傲慢，而非专业。
Detailed Summary详细概述
The Ancestral Baseline
Yudkowsky opens by grounding the problem in evolutionary psychology. Our environment of evolutionary adaptedness (EEA) consisted of hunter-gatherer bands of at most 200 people, with no writing. In such a world, all background knowledge was either strictly private (where you found the oasis) or completely universal (what an oasis is, why water matters). You were essentially never more than one inferential step from any fellow band member. No abstract disciplines existed with bodies of evidence generalized into elegant theories — no inferential chains a hundred steps long.
The Instinct That Didn't Update
This produced a social heuristic perfectly calibrated for the EEA: if someone makes a claim with no obvious support, they are a liar or an idiot. You would never think "maybe this person has well-supported background knowledge that no one in my band has heard of" — because in the ancestral environment, that simply did not happen. Conversely, if something is obvious to you and your listener fails to see it, they are the idiot. And if someone makes an unsupported claim while expecting you to believe it, acting indignant when you don't — they must be crazy.
Combined with the illusion of transparency and self-anchoring (modeling other minds as slightly-modified versions of oneself), this explains why communication between experts and lay audiences — or even between scientists from different disciplines — is legendarily difficult.
The Two-Sided Failure
Yudkowsky identifies a two-sided failure pattern:
- Explainers take one step back when they need to take two or more. They assume the audience already has the background concepts needed to follow the next step.
- Listeners assume things should become visible in one step, when they actually take two or more steps to explain. Any claim that looks unsupported after one step gets categorized as crazy.
Both sides act as if they expect very short inferential distances from universal knowledge to any new knowledge.
The Biologist and the Physicist
The central worked example: a biologist can justify evolution to a physicist by saying it is the simplest explanation. Both have been steeped in the history of science from Newton to Einstein, which invests "simplest explanation" with its full power — a Word of Power, spoken at the birth of theories and carved on their tombstones. But someone without that history hears the phrase as an interesting but hardly decisive argument. To them, "simplest explanation" doesn't feel like a knockdown tool for comprehending office politics or fixing a car.
So the nonscientist concludes the biologist is infatuated with their own ideas, too arrogant to consider alternatives. And the biologist, who has explained it's the simplest explanation, concludes that nonscientists are just idiots.
The Requirements for Clear Argument
A clear argument must:
- Lay out an inferential pathway, starting from what the audience already knows or accepts.
- Never make a statement without obvious justification in arguments already established with that audience.
- Never appear to weight an argument more heavily than is justified in the audience's eyes at that time — even if you know the argument is actually decisive.
- Not signal that you think you are working dozens of inferential steps ahead of your audience. That reads as condescending, not expert.
The essay closes with a recursive joke: if you think you can explain "systematically underestimated inferential distances" briefly, in just a few words — you are yourself exhibiting the very bias the essay describes.
祖先的基准
Yudkowsky 从进化心理学出发奠定问题的根基。我们的进化适应环境（EEA）由最多 200 人组成的狩猎采集族群构成，没有文字。在这样的世界里，所有背景知识要么是严格私有的（你发现绿洲在哪里），要么是完全普遍共享的（绿洲是什么，水为什么重要）。你与任何族群成员之间，本质上从不超过一步的推理距离。彼时不存在任何抽象学科——没有将大量证据归纳为优雅理论的知识体系，没有长达百步的推理链条。
未被更新的本能
这催生了一种对 EEA 完美校准的社会启发：如果有人提出没有明显依据的说法，他要么是说谎者，要么是蠢货。你绝不会想「也许这个人拥有族群里没有人听说过的、有充分支撑的背景知识」——因为在祖先环境中，那根本不会发生。反过来，如果某件事对你显而易见，而听众却看不出来，他们才是蠢货。而如果有人提出没有依据的说法却期待你相信，你不信时还一脸愤慨——那他一定是疯了。
结合透明度错觉与自我锚定（将他人心智建模为自己的略微变体），这解释了为何专家与外行之间的沟通——乃至不同学科的科学家之间的沟通——历来如此困难。
双向的失败
Yudkowsky 识别出一种双向失败模式：
- 解释者在需要退两步甚至更多时，只退了一步。他们假设听众已经具备跟进下一步所需的背景概念。
- 听众假设事物在一步之内就应当显而易见，而实际上需要两步乃至更多步才能解释清楚。任何在一步之后看起来缺乏依据的说法，都会被归入「疯话」一类。
双方都在表现得好像预期：从普遍知识到任何新知识，推理距离都应当很短。
生物学家与物理学家
核心例证：一位生物学家可以通过说「这是最简解释」来向物理学家论证进化论。双方都浸润在从牛顿到爱因斯坦的科学史中，这段历史赋予了「最简解释」以完整的力量——一个权能之词，在理论诞生时道出，在理论的墓碑上镌刻。但没有这段历史的人，只会把这个短语听作一个有趣但绝非决定性的论据。对他们来说，「最简解释」并不感觉像是理解办公室政治或修理汽车的有力工具。
于是，非科学家断定生物学家迷恋自己的想法，傲慢得不肯考虑其他解释。而生物学家——明明已经解释了「这是最简解释」——则断定非科学家只是蠢货。
清晰论证的要求
一个清晰的论证必须：
- 铺陈一条推理的路径，从听众已知或已接受的内容出发。
- 绝不在未与该听众建立的论证基础上做出没有明显依据的陈述。
- 绝不表现得好像某论据的分量比听众此刻所能感受到的更重——即便你知道该论据实际上是决定性的。
- 不要流露出你认为自己比听众领先数十个推理步骤的信号。那会被解读为居高临下，而非专业。
文章以一个自反的笑话收尾：如果你认为可以用寥寥数语简短解释「系统性低估推理距离」——你自己正在展示的，恰恰是这篇文章所描述的那种偏差。
FAQ常见问答
What exactly is an "inferential step"?「推理步骤」究竟是什么意思？
An inferential step is a move from one piece of knowledge to the next that requires some background concept or reasoning. In the ancestral environment, two people sharing information typically needed at most one such step. A modern scientific conclusion may require a hundred — each step depending on concepts the previous step introduced.
推理步骤是从一个知识点迈向下一个知识点的过程，需要某种背景概念或推理。在祖先环境中，两个人分享信息通常最多需要一步。而一个现代科学结论可能需要一百步——每一步都依赖前一步所引入的概念。
Why does the EEA matter — we've had writing and science for millennia?为什么 EEA 很重要——我们已经有文字和科学好几千年了？
Evolutionary timescales vastly outrun cultural ones. The cognitive hardware built for 200-person bands hasn't been redesigned in the few thousand years since writing appeared. So our instinctive social responses to unsupported claims — "liar or idiot" — still fire even in contexts where the ancestral shortcut is completely wrong.
进化的时间尺度远远超过文化的时间尺度。为 200 人族群构建的认知硬件，在文字出现后短短数千年里并没有被重新设计。所以我们对缺乏依据的说法的本能社会反应——「说谎或蠢货」——即便在祖先的那条捷径完全失效的语境下，仍然会自动触发。
How is this different from just saying "know your audience"?这与「了解你的听众」有什么不同？
"Know your audience" is practical advice about tone and vocabulary. Yudkowsky's point is deeper and more pessimistic: even when communicators try to adapt, they systematically underestimate the number of inferential steps they need to bridge. The failure isn't lack of effort — it's a structural cognitive bias toward assuming shared background.
「了解你的听众」是关于语气和措辞的实用建议。Yudkowsky 的观点更深刻，也更悲观：即便沟通者尝试去适应，他们也会系统性地低估需要弥合的推理步骤数量。失败不是因为努力不够——而是一种结构性的认知偏差：默认存在共同背景。
What is "self-anchoring" and why does it make the problem worse?「自我锚定」是什么，为什么会让问题更糟？
Self-anchoring is the tendency to model other minds as slightly modified versions of your own. Combined with the short-inferential-distance bias, it means you assume others have almost the same conceptual background as you — because they're basically you, with minor tweaks. This makes it feel like any gap should be bridgeable in one or two sentences, when it may require pages.
自我锚定是将他人心智建模为自己的略微变体的倾向。与短推理距离偏差相结合，意味着你会假设他人拥有与你几乎相同的概念背景——因为他们基本上就是你，只是稍有调整。这让人感觉任何差距都可以用一两句话弥合，而实际上可能需要好几页。
Does this mean all technical communication should start from first principles?这是否意味着所有技术沟通都应从第一性原理出发？
Not necessarily from absolute first principles, but from whatever shared premises you can actually establish with your audience. The practical move is to deliberately check: have I actually established with this audience the concepts I am now relying on? Starting one inferential step before that point is not enough.
不一定要从绝对的第一性原理出发，而是从你实际上能与听众建立的共同前提出发。实际操作是刻意检查：我现在所依赖的概念，是否已经与这位听众建立起来了？从那个节点前一步开始是不够的。
What is the essay's closing joke really saying?文章结尾的那个笑话究竟在说什么？
The final line — "if you think you can explain 'systematically underestimated inferential distances' briefly, in just a few words, I've got some sad news for you" — is self-referential. The concept itself requires enough background to understand why the ancestral heuristic is miscalibrated, why modern knowledge is structured differently, and what an inferential step even is. It can't be compressed without losing the argument.
最后那句话——「如果你认为可以用寥寥数语解释'系统性低估推理距离'，我有个坏消息要告诉你……」——是自反的。这个概念本身需要足够的背景才能理解：为什么祖先的启发式规则失准了，为什么现代知识的结构不同，推理步骤究竟是什么。它无法被压缩而不失去论证本身。
In-depth Analysis · Pros & Cons深入解读 · 优缺点
This essay proposes an evolutionary-psychological explanation for a widely experienced but rarely diagnosed failure: the gap between how much background a communicator assumes and how much a listener actually has. It is a piece of applied cognitive science — functional, diagnostic, and prescriptive.
这篇文章为一种广泛存在却罕有诊断的失败提出了一个进化心理学解释：沟通者所预设的背景知识量与听众实际拥有的知识量之间的落差。这是一篇应用认知科学的文章——功能性的、诊断性的，也是规范性的。
- Mechanistic explanation, not just description机制性解释，而非单纯描述Rather than just saying "experts communicate badly," Yudkowsky roots the failure in a specific evolutionary mismatch between the EEA and modern knowledge structures. This makes the bias feel inevitable rather than merely a personal shortcoming.Yudkowsky 不是简单地说「专家沟通很差」，而是将这一失败根植于 EEA 与现代知识结构之间一种具体的进化不匹配。这让这种偏差感觉是不可避免的，而非仅仅是个人的缺陷。
- The symmetry of the failure失败的对称性Identifying the problem as two-sided — explainers undershoot and listeners over-pathologize — avoids the usual framing of "it's just bad teachers" or "it's just dumb audiences," and is more honest about why the impasse is stable.将问题识别为双向的——解释者退步不够且听众过度病理化——避免了「只是老师不好」或「只是听众太蠢」的惯常框架，也更诚实地说明了为什么这种僵局是稳定的。
- The biologist example is precisely chosen生物学家例子选得精准"Simplest explanation" is ideal because it is decisive to someone who has internalized Occam's Razor in a scientific context, and it genuinely does sound weak to someone who hasn't. There's no strawmanning: the non-expert's response is rational given their background.「最简解释」是理想的例子，因为它确实对内化了科学语境中奥卡姆剃刀的人具有决定性力量，而对没有这段背景的人来说确实听起来很无力。没有任何稻草人谬误：非专家的反应在其背景下是理性的。
- Prescriptive payoff is clear规范性收益清晰The essay arrives at a concrete directive: lay out a full inferential pathway from actually-shared premises. This is actionable, not just diagnostic.文章最终给出了一个具体的指令：从实际共享的前提出发铺陈完整的推理路径。这是可操作的，而非仅仅是诊断。
- The EEA hypothesis is harder to verify than the essay impliesEEA 假说比文章暗示的更难以验证Using evolutionary psychology to explain a cognitive pattern is a compelling narrative, but the specific claim that our inferential-distance heuristics were calibrated to 200-person bands is speculative. The pattern of underestimation might have alternative explanations (motivated cognition, overconfidence in one's own clarity) that don't require the EEA story.用进化心理学解释认知模式是个引人入胜的叙事，但「我们的推理距离启发式规则是针对 200 人族群校准的」这一具体说法带有推测性质。这种低估模式可能有其他解释（动机性认知、对自身表达清晰度的过度自信），不一定需要 EEA 故事。
- The prescription undersells difficulty处方低估了困难"Lay out a full inferential pathway starting from what the audience already knows" is the right advice but almost impossibly hard to execute well. Knowing what your audience knows requires precisely the kind of perspective-taking this essay says we're bad at. The essay diagnoses the disease but is vague about the medicine.「从听众已知内容出发铺陈完整推理路径」是正确建议，但执行起来近乎不可能做好。知道听众了解什么，恰恰需要这篇文章所说我们不擅长的那种视角采择。文章诊断了疾病，但对药方语焉不详。
- Conflates different types of communication failure混淆了不同类型的沟通失败The essay treats expert-to-layperson failure and expert-to-different-discipline failure as the same phenomenon. But the latter often involves genuine conceptual disagreement, not just missing background — two biologists can fight about natural selection without any inferential gap. Attributing all miscommunication to distance rather than conflict may be too clean.文章将专家对外行的失败与专家对不同学科的失败视为同一现象。但后者往往涉及真正的概念分歧，而非仅仅缺乏背景——两位生物学家可以在没有任何推理距离差距的情况下争论自然选择。将所有沟通失败都归因于距离而非冲突，可能过于整洁。
- No empirical anchor缺乏实证依据The essay is entirely argumentative and anecdotal. The core claim — that people systematically underestimate inferential distances — could in principle be tested (e.g., via studies of how experts predict novice comprehension), but no such evidence is cited. Given that the essay is making a claim about a systematic cognitive bias, this omission is notable.这篇文章完全依靠论证和轶事。核心主张——人们系统性地低估推理距离——在原则上是可以检验的（例如通过研究专家如何预测新手的理解），但没有引用任何此类证据。鉴于文章是在主张一种系统性认知偏差，这个缺失值得注意。
A short, well-structured essay that offers a genuine explanatory hypothesis for a real problem. Its evolutionary framing is speculative but fruitful, and the symmetry it identifies in the failure pattern is genuinely illuminating. The main gap is the prescription: knowing you should bridge the full inferential gap is not the same as knowing how, and that is precisely where the hard work of communication lives.
一篇简短、结构良好的文章，为一个真实问题提供了一个真正的解释性假说。其进化框架带有推测性，但富有启发性；它对失败模式对称性的识别真正令人眼亮。主要的缺口在于处方：知道应该弥合完整的推理距离，与知道如何弥合，是两码事——而恰恰那里才是沟通的真正困难所在。
Original Text原文
Homo sapiens’s environment of evolutionary adaptedness (a.k.a. EEA or “ancestral environment”) consisted of hunter-gatherer bands of at most 200 people, with no writing. All inherited knowledge was passed down by speech and memory.
In a world like that, all background knowledge is universal knowledge. All information not strictly private is public, period.
In the ancestral environment, you were unlikely to end up more than one inferential step away from anyone else. When you discover a new oasis, you don’t have to explain to your fellow tribe members what an oasis is, or why it’s a good idea to drink water, or how to walk. Only you know where the oasis lies; this is private knowledge. But everyone has the background to understand your description of the oasis, the concepts needed to think about water; this is universal knowledge. When you explain things in an ancestral environment, you almost never have to explain your concepts. At most you have to explain one new concept, not two or more simultaneously.
In the ancestral environment there were no abstract disciplines with vast bodies of carefully gathered evidence generalized into elegant theories transmitted by written books whose conclusions are a hundred inferential steps removed from universally shared background premises.
In the ancestral environment, anyone who says something with no obvious support is a liar or an idiot. You’re not likely to think, “Hey, maybe this person has well-supported background knowledge that no one in my band has even heard of,” because it was a reliable invariant of the ancestral environment that this didn’t happen.
Conversely, if you say something blatantly obvious and the other person doesn’t see it, they’re the idiot, or they’re being deliberately obstinate to annoy you.
And to top it off, if someone says something with no obvious support and expects you to believe it—acting all indignant when you don’t—then they must be crazy.
Combined with the illusion of transparency and self-anchoring (the tendency to model other minds as though the were slightly modified versions of oneself), I think this explains a lot about the legendary difficulty most scientists have in communicating with a lay audience—or even communicating with scientists from other disciplines. When I observe failures of explanation, I usually see the explainer taking one step back, when they need to take two or more steps back. Or listeners assume that things should be visible in one step, when they take two or more steps to explain. Both sides act as if they expect very short inferential distances from universal knowledge to any new knowledge.
A biologist, speaking to a physicist, can justify evolution by saying it is the simplest explanation. But not everyone on Earth has been inculcated with that legendary history of science, from Newton to Einstein, which invests the phrase “simplest explanation” with its awesome import: a Word of Power, spoken at the birth of theories and carved on their tombstones. To someone else, “But it’s the simplest explanation!” may sound like an interesting but hardly knockdown argument; it doesn’t feel like all that powerful a tool for comprehending office politics or fixing a broken car. Obviously the biologist is infatuated with their own ideas, too arrogant to be open to alternative explanations which sound just as plausible. (If it sounds plausible to me, it should sound plausible to any sane member of my band.)
And from the biologist’s perspective, they can understand how evolution might sound a little odd at first—but when someone rejects evolution even after the biologist explains that it’s the simplest explanation, well, it’s clear that nonscientists are just idiots and there’s no point in talking to them.
A clear argument has to lay out an inferential pathway, starting from what the audience already knows or accepts. If you don’t recurse far enough, you’re just talking to yourself.
If at any point you make a statement without obvious justification in arguments you’ve previously supported, the audience just thinks you’re crazy.
This also happens when you allow yourself to be seen visibly attaching greater weight to an argument than is justified in the eyes of the audience at that time. For example, talking as if you think “simpler explanation” is a knockdown argument for evolution (which it is), rather than a sorta-interesting idea (which it sounds like to someone who hasn’t been raised to revere Occam’s Razor).
Oh, and you’d better not drop any hints that you think you’re working a dozen inferential steps away from what the audience knows, or that you think you have special background knowledge not available to them. The audience doesn’t know anything about an evolutionary-psychological argument for a cognitive bias to underestimate inferential distances leading to traffic jams in communication. They’ll just think you’re condescending.
And if you think you can explain the concept of “systematically underestimated inferential distances” briefly, in just a few words, I’ve got some sad news for you . . .
智人的进化适应环境（又称 EEA 或「祖先环境」）由最多 200 人组成的狩猎采集族群构成，没有文字。所有继承的知识都通过言语和记忆代代相传。
在那样的世界里，所有背景知识都是普遍知识。所有非严格私有的信息就是公共信息，仅此而已。
在祖先环境中，你与任何其他人之间，不可能相差超过一个推理步骤。当你发现一处新的绿洲，你不必向同族成员解释绿洲是什么，或者为什么喝水是个好主意，或者怎么走路。只有你知道绿洲在哪里；这是私有知识。但每个人都有理解你对绿洲描述的背景，以及思考水所需的概念；这是普遍知识。在祖先环境中解释事物时，你几乎从不需要解释你的概念。你最多需要解释一个新概念，而不是同时解释两个或更多。
在祖先环境中，不存在任何抽象学科——没有通过书面书籍传播的、将大量精心收集的证据归纳为优雅理论的知识体系，其结论与普遍共享的背景前提之间相距一百个推理步骤。
在祖先环境中，任何说出没有明显依据的东西的人，都是说谎者或蠢货。你不可能想到「嘿，也许这个人有我们族群里没有人听说过的、有充分支撑的背景知识」，因为这在祖先环境中是一个可靠的不变量——那根本不会发生。
反过来，如果你说出某个显而易见的东西，而对方却看不出来，他们才是蠢货，或者是在故意顽固地惹你不快。
更有甚者，如果有人说出没有明显依据的话，却期待你相信他——你不信时他还一脸愤慨——那他一定是疯了。
结合透明度错觉与自我锚定（将他人心智建模为自己的略微变体的倾向），我认为这在很大程度上解释了大多数科学家在与外行沟通——乃至与其他学科的科学家沟通——时那举世皆知的困难。当我观察解释失败时，我通常看到解释者退了一步，而他们需要退两步或更多步。或者听众假设事物在一步内就应当显而易见，而实际上需要两步或更多步来解释。双方的表现都好像预期：从普遍知识到任何新知识，推理距离都应当很短。
一位生物学家对物理学家说，可以通过说它是最简解释来论证进化论。但并非地球上所有人都被灌输过那段传奇的科学史——从牛顿到爱因斯坦——那段历史赋予了「最简解释」以其令人肃然起敬的意义：一个权能之词，在理论诞生时道出，在理论的墓碑上镌刻。对其他人来说，「但这是最简解释！」听起来可能是个有趣却绝非决定性的论据；它感觉上并不像是理解办公室政治或修理汽车的那么有力的工具。显然，生物学家迷恋自己的想法，傲慢得不肯接受那些听起来同样有道理的替代解释。（如果对我来说听起来有道理，它就应该对我族群里任何神志清醒的成员都有道理。）
而从生物学家的角度来看，他们能理解进化论起初听起来可能有些奇怪——但当有人在生物学家解释了进化论是最简解释之后仍然拒绝接受时，那显然非科学家就是蠢货，和他们说话毫无意义。
一个清晰的论证必须铺陈出一条推理的路径，从听众已知或已接受的内容出发。如果你回溯得不够远，你就只是在自言自语。
如果在任何一个节点，你做出了一个在你先前已支撑的论证中没有明显依据的陈述，听众只会认为你疯了。
当你允许自己被看到明显地给某个论据附加超出听众此刻在其眼中所能判断的分量时，也会发生这种情况。例如，表现得好像你认为「更简单的解释」是支持进化论的决定性论据（事实上确实如此），而非一个还算有趣的想法（对没有从小被灌输敬畏奥卡姆剃刀的人来说，听起来就是如此）。
哦，你最好不要流露出任何暗示——表明你认为你正在距听众所知的东西十几个推理步骤之外工作，或者你认为自己拥有他们无法获得的特殊背景知识。听众对「系统性低估推理距离导致沟通堵塞这一认知偏差的进化心理学论证」一无所知。他们只会认为你在居高临下。
如果你认为你可以用寥寥数语、简短地解释「系统性低估推理距离」这个概念，我有个坏消息要告诉你……