August 13, 1961, before dawn. East German workers began laying barbed wire across Berlin. Within days the wire had become concrete. A city split in two.
The Berlin Wall is usually read as a symbol of the Cold War. But it is also evidence of something more specific: a system that required physical means to stop people from leaving had answered, on its own behalf, the question of what it was.
You don't build a wall to stop people from flowing in. You build a wall to stop people from escaping.
In the twelve years before construction, East Germany had lost 3.5 million citizens — roughly one-fifth of its total population. These people were not expelled or exiled. They left on their own. They voted with their feet.
A system that can only function by locking people inside has already answered the question "are people a purpose or a tool in this system?"
To be fair: the Soviet model did not fail in all respects. In 1957 the Soviet Union launched the first artificial satellite, shocking the world. In 1961, Gagarin became the first human in space. Soviet nuclear weapons, ballet, chess players, Olympic teams — all of world-class quality. From near-zero to industrialization, from agricultural nation to superpower, the Soviet Union achieved in under thirty years a transformation unprecedented in human history.
Closed systems have extraordinary efficiency at concentrating resources on a single goal. When you don't need anyone's consent, don't need to survive any debate, don't need to accommodate any dissent, you can focus an entire nation's total force like a laser onto a single point.
The cost: everything not in focus was deteriorating. The Soviet Union could build the best rockets but couldn't manufacture shoes that fit comfortably. Could train world champions but made ordinary people queue for hours for bread. Could design nuclear warheads but couldn't design an administrative system that allowed low-level officials to honestly report actual data.
Because the closed system's efficiency came precisely from severing feedback loops. No dissent, no friction, no efficiency losses. But also no error correction. Errors were not repaired — they were concealed, redefined, or blamed on "saboteurs." Khrushchev tried limited crack-opening. His 1956 "Secret Speech" disclosed some of Stalin's crimes. But he quickly discovered the problem: once a crack is opened, it is very difficult to control how wide it grows. Hungarians thought the crack meant freedom and revolted. The Soviet response was to send tanks to close the crack.
1968: Czechoslovakia's Prague Spring tried to create "socialism with a human face" — essentially chiseling more cracks into the Soviet system's closure. The Soviet Union sent tanks again.
These two events established an unwritten rule: the width of any crack was for the top to decide. Widening a crack from below was an invitation for tanks.
The same period: what was the United States doing? A complete mess. McCarthyism branded half of Hollywood and a generation of academics as "communist sympathizers." Vietnam swallowed 58,000 American lives and millions of Vietnamese. Watergate revealed a sitting president openly lying and abusing power. Racial segregation was only legally abolished in the 1960s, with actual discrimination persisting for decades beyond that. Urban riots, political assassinations, stagflation — the America of the 1960s and 1970s looked like a country coming apart.
If you took a snapshot in 1970 and compared the two systems, the Soviet side appeared stable, predictable, uniformly ordered. The American side appeared chaotic, fractured, consumed by self-doubt. Many observers at the time — including many Western intellectuals — genuinely believed the Soviet model was more efficient and might represent the future.
They saw efficiency. They did not see resilience.
McCarthy was eventually challenged before a live television audience by one of his own colleagues: "Senator, have you no sense of decency?" His political career subsequently collapsed — not because some center of power decided to discard him, but because pluralistic information channels let the public see his true face.
Vietnam generated mass anti-war movements at home. Independent press coverage brought the front's realities into every American living room. The government could not control this information — it tried, and failed. The United States eventually withdrew from Vietnam. The withdrawal was humiliating. But the system received the feedback that "this path was wrong" and made a corresponding adjustment.
Watergate led to Nixon's resignation — a sitting president driven from office by his own nation's judiciary and press. This was unimaginable in the Soviet Union, not because Soviet leaders didn't make mistakes, but because the Soviet system had no force capable of holding leaders accountable.
Every American crisis was ugly, noisy, the laughingstock of the world. But after every crisis, the system made some kind of adjustment. Not because the system was "good," but because the system's foundational structure contained a few things: information channels capable of transmitting bad news, institutional arrangements capable of challenging power, legitimate pathways capable of expressing dissatisfaction.
These things made the system look clumsy. Democratic decision-making is slower than autocracy, debate is less efficient than command, compromise is messier than coercion. But behind the clumsiness was a capacity — a capacity to walk out alive from its own mistakes. The Soviet system looked efficient. But behind the efficiency was an incapacity — an incapacity to even recognize its own errors, let alone repair them.
November 9, 1989. The Berlin Wall came down. But it was not pushed down. Not stormed by an army. Not destroyed by a revolution. It simply fell when, following a press conference misstatement, ordinary people walked through it.
East German government spokesman Schabowski announced new travel regulations at a press conference. A reporter asked: when does this take effect? He shuffled through his papers, and said uncertainly: "As far as I know... immediately."
That night, tens of thousands of East Berliners surged toward the checkpoints. The border guards didn't know what to do — no one had ordered them to shoot, and no one had ordered them to let people through. Finally they stepped aside. People walked across.
The absurdity of this scene is meaningful. A system that had controlled tens of millions of people for forty years was not defeated in an epic confrontation. It evaporated in a small information confusion.
Why? Because by 1989, the Soviet model's legitimacy was an empty shell. The Soviet system's legitimacy was built on a temporal promise: history is moving toward communism; every sacrifice is for that destination. But by the 1980s, no one believed this promise — including most people inside the system. They continued operating as the system required not because they believed, but because of inertia.
Gorbachev's reforms attempted to do one thing: maintain the system while introducing openness. Let people tell the truth, let information circulate, let feedback loops rebuild. He thought this would make the system stronger. The opposite happened. When a system built on closure is allowed to open, it does not "get better." It exposes everything that closure had been concealing — corruption, inefficiency, lies, hollow promises. And once those things are seen, the system's legitimacy is not merely weakened. It evaporates entirely.
Because this system's legitimacy was not built on "I have made your life good" — that at least could be debated. It was built on "history is on our side" — a belief. Once a belief is permitted to be doubted, there is no "partial doubt." Either the belief holds or it vanishes completely.
So the Wall's fall was not the endpoint of a gradual process. It was a single instant. Before that instant, the system appeared to be functioning. After that instant, everyone could see it had actually stopped long ago — but no one had dared say so aloud.
1991: the Soviet Union formally dissolved. Francis Fukuyama declared the end of history. Liberal democracy and market economics had won — not temporarily, but finally and completely.
Reading this series from the beginning, you can see where this argument went wrong.
The Cold War's end proved one thing: a completely closed system — no informational pluralism, no checks on power, no feedback loops — will eventually shatter under the weight of its own rigidity. No matter how strong its industrial capacity, military force, or ideological coherence, if it cannot recognize and repair its own errors, it is counting down to collapse.
But the Cold War's end did not prove another thing: that a "relatively open" system will automatically and continuously treat people as ends.
The existence of cracks makes error correction possible. But cracks do not automatically widen, and correction does not automatically happen. Every correction — the Civil Rights Movement, the anti-war movement, the accountability of Watergate — was the result of people fighting within those cracks. Institutions provided the cracks. But action was still required to let the light through.
Fukuyama's error was equating "the existence of cracks" with "the problem's resolution." As if a system that has cracks — elections, courts, press freedom — will automatically move toward treating people as ends. The past thirty years have demonstrated how naive this was.
The post-Cold War world did not become what Fukuyama imagined. It became something else — a harder-to-recognize form of instrumentalization.
Without the Soviet Union as a mirror, the winning system lost its reference point and its self-vigilance. When no opponent is reminding a system "look at your own problems," the system increasingly operates according to its own internal logic rather than the principles it claims to embody.
The neoliberal globalization of the 1990s elevated market efficiency to near-unquestionable ultimate status. If a policy was "good for the market," it was good. If a person was "useful to the market," they had value. This is structurally identical to Soviet logic — only the Soviet ultimate standard was "historical progress" while the neoliberal ultimate standard is "economic efficiency." Different ultimate standards; same instrumentalizing logic.
The 2008 financial crisis was a perfect case. Every actor in the system — bankers, ratings agencies, regulators, homebuyers — was doing "rational" things by the system's rules. The system's overall output was catastrophe. Afterward, almost no one was held accountable. Taxpayer money was used to rescue the institutions that had created the crisis. What did the ordinary people who lost their homes receive?
This scene is structurally identical to the appeasement policy analyzed in the second essay: the system maintained itself; the cost was borne by specific people.
Then came social media and the algorithmic age. Information channels were formally more pluralistic than ever — anyone could speak. But algorithms determined whose voice was heard. Platforms were optimized not for "giving people true and useful information" but for "keeping people on the platform longer." Anger retains attention better than reason; the extreme generates more engagement than the moderate. The result: information channels formally expanded while information quality was substantively degraded.
The cracks appeared to remain. But the light coming through had become murky.
Five essays: from the railway car of 1918 to the Berlin Wall of 1989 to today.
If this series has a central message, it is: people becoming ends is never granted by institutions, never guaranteed by declarations, never achieved by one system defeating another.
People becoming ends happens in specific people, at specific moments, under specific pressures, choosing not to treat themselves or others as means.
The best thing institutions can do is not to declare "people are ends" — such declarations are too cheap. The best thing institutions can do is to keep the cracks open: let different voices have the chance to be heard, let power remain challengeable, let errors have channels through which to be recognized, let the question "who counts as human" never be permanently closed.
These cracks will not widen on their own. They need to be fought for, maintained, used. Unused cracks narrow. When they narrow enough, the system becomes once again a machine concerned only with its own operation, and the people inside it become components once more.
In the ruins of 1945, humanity chiseled open a few cracks. Eighty years later, some are widening and some are narrowing. Which direction they ultimately take depends on whether each person alive today still feels this has something to do with them.
1961年8月13日凌晨,东德工人开始在柏林铺设铁丝网。几天之内,铁丝网就变成了混凝土墙。一座城市被一道墙劈成两半。
柏林墙通常被理解为冷战的象征。但它其实还说明了一件更具体的事:一个系统需要用物理手段阻止人离开,这本身就是一份供词。没有人需要用墙来阻止人涌入。墙是用来阻止人逃走的。
东德政府在修墙之前的十二年里,失去了三百五十万公民——大约是全国人口的五分之一。这些人不是被驱逐,不是被流放。他们是自己走的,用脚投的票。
一个系统如果需要把人关在里面才能运转,那它自己已经回答了"人在这个系统里是目的还是工具"这个问题。
公平地说,苏联模式不是在所有方面都"失败"的。1957年,苏联把第一颗人造卫星送上太空,震惊了全世界。1961年,加加林成为第一个进入太空的人。苏联的核武器、芭蕾舞团、国际象棋选手、奥运代表团,都是世界顶级水平。从零到工业化、从农业国到超级大国,苏联用了不到三十年,速度是人类历史上前所未有的。
封闭系统在集中资源完成单一目标方面有惊人的效率。当你不需要征求任何人的意见、不需要经过任何辩论、不需要照顾任何异议,你可以把一个国家的全部力量像激光一样聚焦到一个点上。
代价是什么?代价是所有没被聚焦到的地方都在衰败。苏联能造出最好的火箭,但造不出让普通人穿着舒服的鞋子。能训练出世界冠军,但让普通人排几个小时的队才能买到面包。能设计核弹头,但设计不出一个让基层官员敢于汇报真实数据的行政系统。
因为封闭系统的效率恰恰来自于它切断了反馈回路。没有异议,就没有摩擦,就没有效率损失。但也没有纠错。错误不是被修复,而是被掩盖、被重新定义、或者被归咎于"破坏分子"。
赫鲁晓夫试过有限度地打开一些裂缝。1956年他做了那个著名的"秘密报告",揭露了斯大林时代的部分罪行。但他很快发现一个问题:裂缝一旦打开,就很难控制它打开多大。匈牙利人以为裂缝意味着自由,于是起义了。苏联的回应是派坦克进去把裂缝堵上。
1968年,捷克斯洛伐克的"布拉格之春"试图创造一种"有人性面孔的社会主义"。但那本质上就是在苏联模式的封闭系统里凿出更多的裂缝,于是苏联再次派坦克碾了过去。
这两个事件建立了一条不成文的规则:裂缝只能由上面决定开多大。从下面自行扩大裂缝,等同于邀请坦克。
同一时期,美国在做什么?在搞得一团糟。麦卡锡主义把半个好莱坞和大批学者打成了"共产主义同情者"。越战的泥潭吞噬了五万八千条美国人的命和几百万条越南人的命。水门事件暴露了总统在公然说谎和滥用权力。种族隔离持续到1960年代才被法律废除,而法律废除之后的实际歧视又持续了几十年。城市暴乱、政治暗杀、经济滞胀——整个六七十年代的美国看起来像一个正在裂开的国家。
如果在1970年做一个快照,把美苏两个系统放在一起比较,苏联那边是稳定的、可预测的、整齐划一的,美国这边是混乱的、撕裂的、充满了自我怀疑。很多当时的观察者——包括不少西方知识分子——真心认为苏联模式更有效率,可能代表了未来。
他们看到了效率,没有看到韧性。
麦卡锡最终被他自己的同僚在电视直播中质问:"参议员先生,难道你一点廉耻都没有了吗?"于是他的政治生涯随后崩塌。不是因为某个权力中心决定抛弃他,而是因为多元的信息渠道让公众看到了他的真实面目。
越战在国内引发了大规模的反战运动,媒体的独立报道让前线的真实状况进入了每个美国家庭的客厅。政府无法控制这些信息——它不是没试过,但都失败了。最终美国撤出了越南。撤退是屈辱的,但系统接收到了"这条路走错了"的反馈,并且做出了对应的调整。
水门事件导致了尼克松辞职,一个在任总统居然被自己国家的司法和媒体系统逼退。这在苏联是不可想象的,当然不是因为苏联领导人不会犯错,而是因为苏联系统中没有任何力量可以追究领导人的错误。
每一次美国的危机都是丑陋的、喧闹的、让全世界看笑话的。但每一次危机之后,系统都做出了某种程度的调整。不是因为系统"善良",而是因为系统的底层结构中有几样东西:能传递坏消息的信息渠道,能挑战权力的制度安排,能表达不满的合法途径。
这些东西让系统看起来很笨拙。民主的决策过程比独裁慢,辩论比命令低效,妥协比强制难看。但笨拙的背后是一种能力——一种活着走出自己犯下错误的能力。苏联的系统看起来很高效。但高效的背后是一种无能——一种死都不会识别自己错误的无能,遑论如何修复这些错误了。
1989年11月9日,柏林墙倒了。但它不是被推倒的,不是被军队攻克的,不是被革命摧毁的。它只是在一场新闻发布会的口误之后,被一群普通人走过去的。
东德政府发言人沙博夫斯基在记者会上宣布了新的旅行规定。一个记者问:什么时候生效?他翻了翻手里的文件,不太确定地说:"据我所知……立即生效。"
当晚,成千上万的东柏林人涌向边境检查站。边防军不知道该怎么办——没有人给他们下过开枪的命令,也没有人给他们下过放行的命令。最后他们让开了,然后人群就这样走了过去。
这个场景的荒诞感是有深意的。一个控制了几千万人长达四十年的系统,不是在一场史诗般的对决中被击败的。它是在一个小小的信息混乱中蒸发的。
为什么?因为到了1989年,苏联模式的正当性已经是一个空壳了。苏联系统的正当性建立在一个时间性的承诺上:历史正在走向共产主义,一切牺牲都是为了那个终点。但到了八十年代,没有人再相信这个承诺了——包括系统内部的大多数人。他们继续按照系统的要求行事,不是因为相信,而是因为惯性。
戈尔巴乔夫的改革试图做一件事:在保留系统的同时引入开放性。让人们可以说真话,让信息可以流通,让反馈回路重新建立。他以为这会让系统变得更强。结果恰恰相反,当一个建立在封闭性之上的系统被允许打开时,它不会"变好"。它会暴露出封闭所掩盖的一切,包括腐败、低效、谎言、空洞的承诺。而一旦这些东西被看到,系统的正当性不是被削弱而已,是直接蒸发了。
因为这个系统的正当性不是建立在"我让你过上了好日子"这样的经验事实上——那至少还可以争论——而是建立在"历史站在我们这边"这样的信念上。信念一旦被允许怀疑,就不存在"部分怀疑",于是信念要消失,就会直接整体消失。
所以柏林墙的倒塌不是一个渐进过程的终点,它是一个瞬间。在那个瞬间之前,系统看起来还在运转。在那个瞬间之后,所有人都看到了它其实早就停了,只是没人敢说破而已。
1991年,苏联正式解体。弗朗西斯·福山写下了那个著名的论断:历史终结了。自由民主制和市场经济赢了。不是暂时地赢了,是终极地赢了。
如果读过这个系列的前四篇,应该能看出这个论断的谬误在哪里。
冷战的结束证明了一件事:一个完全封闭的系统——没有信息多元性,没有权力制衡,没有反馈回路——最终会因为自身的刚性而碎裂。不管它的工业能力多强、军事力量多大、意识形态多完整,如果它不能识别和修复自己的错误,它就在倒计时。
但冷战的结束没有证明另一件事:一个"比较开放"的系统就会自动地、持续地把人当作目的。
裂缝的存在让纠错成为可能。但裂缝不会自动扩大,纠错不会自动发生。每一次纠错——民权运动、反战运动、水门事件的追责——都是人在裂缝中斗争的结果。制度提供了裂缝,但还需要有人去行动,才能让光从裂缝中透进来。
福山的错误在于把"裂缝的存在"等同于"问题的解决"。好像只要一个系统有裂缝——有选举、有法院、有新闻自由——它就会自动走向"人是目的"的状态。过去三十年的历史证明了这有多天真。
冷战结束后的世界没有变成福山预想的样子。它变成了另一种东西——一种更难辨认的工具化形态。
没有了苏联这面镜子,胜出的那个系统失去了参照点,也失去了自我警惕。当没有对手在提醒一个系统"看看你自己的问题"时,系统就会越来越多地按照自身的逻辑运转,而不是按照它声称的原则运转。
九十年代的新自由主义全球化把市场效率提升为近乎不可质疑的终极标准。如果一个政策"对市场好",它就是好的。如果一个人"对市场有用",他就有价值。这和苏联的逻辑在结构上是一样的——只不过苏联的终极标准是"历史进步",而新自由主义的终极标准是"经济效率"。终极标准不同,工具化的逻辑相同。
2008年金融危机是一个完美的案例。系统中的每一个参与者——银行家、评级机构、监管者、购房者——都在按照系统的规则做"理性"的事。系统整体的输出是灾难。危机之后,几乎没有人被追究。纳税人的钱被用来拯救制造了危机的机构。那些失去了房子的普通人得到了什么?
这一幕和第二篇分析的绥靖政策在结构上如出一辙:系统维护了自己,代价由具体的人来承担。
然后是社交媒体和算法时代。信息渠道在形式上比以往任何时候都更多元——任何人都可以发言。但算法决定了谁的声音被听到。平台的优化目标不是"让人获得真实有用的信息",而是"让人花更多时间在平台上"。愤怒比理性更能留住注意力,极端比温和更能制造互动。结果是:信息渠道在形式上扩大了,但信息质量在实质上被侵蚀了。
裂缝看起来还在,但透进来的光变浑浊了。
这正是当下最隐蔽的危险:裂缝没有被封住——没有柏林墙,没有审查机关,没有秘密警察——但裂缝的功能在退化。信息多元但不可靠,选举存在但选民对立,抗议是合法的但很少真的改变什么。系统没有变成苏联。它变成了一种新的东西——一种保留着开放性的外观但逐渐丧失开放性内容的东西。
五篇文章走到这里,从1918年的那节火车车厢,到1989年的柏林墙,到今天。
如果这个系列有一个核心信息,那就是:人成为目的,从来不是制度赐予的,也不是宣言保证的,更不是某一个系统打败另一个系统就能实现的。
人成为目的这件事,发生在具体的人,在具体的时刻,顶着具体的压力,选择不把自己和别人当作工具的那个瞬间。
制度能做的最好的事情,不是宣布"人是目的"——这种宣布太廉价了。制度能做的最好的事情是保持裂缝的存在:让不同的声音有机会被听到,让权力有可能被质疑,让错误有渠道被识别,让"谁算是人"这个问题永远不被关上。
这些裂缝也不会自动拓宽。它们需要被人争取、被人维护、被人使用。不被使用的裂缝会收窄。当收窄到一定程度,系统就又变成了一台只关心自身运转的机器,里面的人又变成了零件。
1945年的废墟中,人类凿开了几条裂缝。八十年后,这些裂缝有的在扩大,有的在收窄。它们最终走向哪个方向,取决于今天的每一个人是否还觉得这件事和自己有关。